August 2026: 846 archived news items, newest first. Items here are older than the 24-hour Latest News window; the live feed is at /news/.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: mastracode@0.34.0. RepoRadar is keeping the source link direct for verification.
This lands on mastra-ai/mastra - gauge whether it shifts what you already ship. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: mastra@1.25.1. RepoRadar is keeping the source link direct for verification.
At its core this is about mastra-ai/mastra, worth a look if that's in your workflow. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: @mastra/vercel@1.4.0. RepoRadar is keeping the source link direct for verification.
What's actually new here is mastra-ai/mastra - see if it moves anything you maintain. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: @mastra/vectorize@1.1.1. RepoRadar is keeping the source link direct for verification.
Builders evaluating mastra-ai/mastra should verify the source before changing a production default. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: @mastra/upstash@1.4.1. RepoRadar is keeping the source link direct for verification.
This lands on mastra-ai/mastra - gauge whether it shifts what you already ship. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: @mastra/temporal@0.3.2. RepoRadar is keeping the source link direct for verification.
The thing to notice is mastra-ai/mastra; decide whether it changes your next build. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: @mastra/tanstack-start@0.2.16. RepoRadar is keeping the source link direct for verification.
What's actually new here is mastra-ai/mastra - see if it moves anything you maintain. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: @mastra/spanner@1.6.1. RepoRadar is keeping the source link direct for verification.
What's actually new here is mastra-ai/mastra - see if it moves anything you maintain. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: @mastra/server@1.60.0. RepoRadar is keeping the source link direct for verification.
The thing to notice is mastra-ai/mastra; decide whether it changes your next build. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
GitHub Releases published a source-backed AI update around mastra-ai/mastra: @mastra/sentry@1.2.11. RepoRadar is keeping the source link direct for verification.
This lands on mastra-ai/mastra - gauge whether it shifts what you already ship. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
arXiv:2607.01854v2 Announce Type: replace-cross Abstract: Can a platform tell, before deployment, whether an open-weight checkpoint has had its refusal mechanism stripped? Runtime guards cannot: they score generations, not the artifact.
This lands on Has This Checkpoint Been Abliterated? A Two-Signal Audit - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2607.26654v3 Announce Type: replace-cross Abstract: Post-training alignment is often shallow, eroding under fine-tuning. It remains untested as to whether constitutional midtraining interventions can produce durable alignment when cleanly isolated from post-training.
Builders evaluating Constitutional Midtraining should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.11210v2 Announce Type: replace Abstract: Bayesian calibration of process-based models requires a prior distribution for each model parameter. Despite decades of methodological work, researchers almost always fall back on uniform priors.
Operators using related systems should check whether Distribird changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16373v2 Announce Type: replace-cross Abstract: Despite comprising over 70% of its surface, the world's oceans are critically underobserved compared to the land surface or the atmosphere. Understanding the global ocean requires jointly observing its surface and subsurface structure, yet no standardized, high
Operators using related systems should check whether OceanDepths changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16956v1 Announce Type: new Abstract: API buyers purchase a dated contract, not a model name alone: the contract includes the requested and served model, reasoning-effort term or its omission, output rail, service product, prompt, and price schedule. We study the reasoning-effort term through a registered pai
The thing to notice is The Price of Thinking; decide whether it changes your next build. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.17067v1 Announce Type: new Abstract: As text-to-image generative models advance, they raise critical safety concerns, particularly the generation of Not-Safe-For-Work (NSFW) content such as violence and nudity, further exacerbated by red-teaming adversarial attacks. Existing defenses predominantly operate un
At its core this is about DiSCO, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.17071v1 Announce Type: new Abstract: We present KernelArc, a multi-agent framework for autonomous GPU kernel optimization across heterogeneous workloads. Strategy-specialized agents run in parallel and coordinate through conclusions-only shared memory, a deterministic benchmark guard, and read-only cross-age
What's actually new here is KernelArc - see if it moves anything you maintain. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.17183v1 Announce Type: new Abstract: Small Language Models (SLMs) are increasingly deployed in resource-constrained, privacy-sensitive settings, where safety and bias failures can cause security and societal risks. However, existing AI safety\slash security\slash compliance benchmarks are designed for large
What's actually new here is Benchmarking the Benchmarks - see if it moves anything you maintain. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.17336v1 Announce Type: new Abstract: Long-context prefill in large language models (LLMs) incurs substantial computation and memory traffic because dense self-attention computes quadratic query-key scores. Existing methods either use a uniform low-precision path or select token interactions, leaving spatial
What's actually new here is TileMix - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.17947v1 Announce Type: new Abstract: Large language models can generate executable programs, which makes it possible to search directly over procedural content generators rather than individual levels. We study this approach in Sokoban, Zelda, Dangerous Dave, and Lode Runner.
Operators using related systems should check whether Procedural Content Metageneration via Program Search and Continual changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model...
arXiv:2311.06273v1 Announce Type: cross Abstract: The rise of ChatGPT has brought a notable shift to the AI sector, with its exceptional conversational skills and deep grasp of language. Recognizing its value across different areas, our study investigates ChatGPT's capacity to predict stock market movements using only
This lands on Potential of ChatGPT in predicting stock market trends - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2602.14784v1 Announce Type: cross Abstract: Breaking long documents into smaller segments is a fundamental challenge in information retrieval. Whether for search engines, question-answering systems, or retrieval-augmented generation (RAG), effective segmentation determines how well systems can locate and return r
At its core this is about Intent-Driven Dynamic Chunking, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16630v1 Announce Type: cross Abstract: Repository-scale coding requires an agent to keep tests, imports, configuration, and migration rules consistent within a bounded context window. We model this as reconstructing a coupled-fact graph: at each edit, a required fact comes from recent context or parametric m
This lands on The Working Set of a Coding Agent - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16897v1 Announce Type: cross Abstract: Large-scale urban simulation plays a pivotal role in social science, traffic safety, and transportation policy. Recent work has shown that large language models, when prompted as agents, can generate lifelike daily routines at city scale.
This lands on CityReal - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16910v1 Announce Type: cross Abstract: National AI strategies increasingly guide governance, workforce development, innovation, and competitiveness, but less is known about how they frame education as a sector with pedagogical, cultural, ethical, and implementation demands. This study develops and applies an
Teams affected by Education-centered critical policy analysis of AI need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.17148v1 Announce Type: cross Abstract: A personal language agent learns a fact from one audience and may later place it in the prompt it assembles for another. This memory-to-context step is an attack surface: ambiguous or inconsistent channels, cross-audience prying, and poisoned memory can each cause the s
Builders evaluating Authorization Before Context should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.17176v1 Announce Type: cross Abstract: An AI audit record is useful only if its durability and trust boundary are explicit. Returning a guarded decision before any durable write minimizes latency, but it cannot guarantee that evidence survives an immediate crash.
Operators using related systems should check whether The Acknowledgment Point Is the System changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.17373v1 Announce Type: cross Abstract: Sample efficiency is a central challenge in reinforcement learning (RL), particularly in image-based domains where agents must learn from high-dimensional visual inputs. Traditional sampling often relies on random or suboptimal experience selection, leading to redundant
Teams affected by Integrating Novelty and Surprise for Experience Prioritization need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent...
arXiv:2608.17948v1 Announce Type: cross Abstract: Recent research has leveraged Large Language Models (LLMs) to enhance Automated Feature Engineering (AutoFE) through semantic descriptions and trajectory-based prompting. However, there exist two challenges that limit their applicability and scalability in long-horizon
This lands on SIGMA - gauge whether it shifts what you already ship. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2510.08713v3 Announce Type: replace Abstract: Enabling embodied agents to imagine future states is essential for robust and generalizable visual navigation. Yet, state-of-the-art systems typically rely on modular designs that decouple navigation planning from visual world modeling, which often induces state-actio
Teams affected by Towards Unified World Models for Visual Navigation need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2605.16024v2 Announce Type: replace Abstract: Desktop GUI agents operate under partial observability: visually similar screens can correspond to different underlying workflow states, so locally plausible actions can lead to sharply different outcomes. We frame this as a problem of computer/OS state exploration, w
Teams affected by ScreenSearch need to decide whether its documented change alters their current workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2311.09614v5 Announce Type: replace-cross Abstract: This study addresses critical gaps in automated lymphoma segmentation from PET/CT images, focusing on issues often overlooked in existing literature. While deep learning has been applied for lymphoma lesion segmentation, few studies incorporate out-of-distributi
This lands on Comprehensive framework for evaluation of deep neural networks - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2406.01586v4 Announce Type: replace-cross Abstract: Diffusion models have been verified to be effective in generating complex distributions from natural images to motion trajectories. Recent diffusion-based methods show impressive performance in 3D robotic manipulation tasks, whereas they suffer from severe runti
What's actually new here is ManiCM - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2604.26703v2 Announce Type: replace-cross Abstract: Scientific discovery often begins where observation and prediction disagree. As computation and machine learning survey chemical space, experiment-simulation mismatches are exposed at scale, while tracing them to physical mechanisms remains expert-led.
At its core this is about Discovering physical mechanisms from experiment-simulation mismatches, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2607.09520v2 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) are the perceptual backbone of embodied AI, but their energy footprint on edge hardware remains poorly understood. Existing efficiency efforts focus predominantly on reducing visual tokens, implicitly treating visual processing as t
Teams affected by Seeing is Free, Speaking is Not need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.04772v2 Announce Type: replace-cross Abstract: Scaling supervision for multi-turn medical agents is difficult because expert dialogue annotation is costly and clinical conversations are privacy-restricted. We introduce Guideline-as-Oracle (GAO), which compiles American Academy of Ophthalmology guidance into
Builders evaluating Guideline-as-Oracle should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
GitHub Copilot for JetBrains now supports enterprise managed settings for plugin governance, MCP server access, OpenTelemetry, and permission modes. Administrators can now apply consistent controls for everyone on your enterprise’s...
Operators using related systems should check whether Enterprise managed settings in GitHub Copilot for JetBrains changes compatibility, cost, or access requirements. For builders, this can change a default in the review, editor, or build workflow you touch...
ChatGPT Ads is expanding to 31 European markets. Learn how advertisers can reach people as they explore, compare options, and make decisions.
What's actually new here is ChatGPT Ads expands across Europe - see if it moves anything you maintain. For builders, a release like this shifts the latency, capability, and cost trade-offs of what to run next.
Building on our self-service credential revocation experiences for incident response, you can now take token-type and user-specific actions to deauthorize and revoke user credentials during a security incident. This gives...
At its core this is about Credential revocation and deauthorization by token type, worth a look if that's in your workflow. For builders, this can change a default in the review, editor, or build workflow you touch every day.
What's changed Added an optional spellcheck setting that underlines misspelled words in the prompt input as you type, using your installed aspell , hunspell , or ispell Fixed whole-prompt-cache invalidation when a language server disconnected or reconnected mid-session Fixed nested markdown list items misaligning at de
At its core this is about anthropics/claude-code, worth a look if that's in your workflow. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
OpenAI launches an initiative to strengthen democratic oversight of AI in national security, supporting government institutions with tools, training, and expertise.
What's actually new here is Strengthening democratic oversight in national security - see if it moves anything you maintain. For teams giving agents real access, this surfaces a failure mode worth weighing before widening agent or tool permissions.
Gemini in Chrome is now available to all Android users in the U.S.
Builders evaluating Tap into the power of Gemini in Chrome should verify the source before changing a production default. For builders, a release like this shifts the latency, capability, and cost trade-offs of what to run next.
How Much Memory Does Your Agent Actually Need?
Builders evaluating How Much Memory Does Your Agent Actually Need? should verify the source before changing a production default. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
Changes since langchain-openai==1.5.1 release(openai): 1.5.2 ( #39719 ) fix(openai): preserve reasoning item boundaries ( #39278 ) release(openai): 1.5.2a1 ( #39709 ) feat(openai): extract gateway metadata from response headers when available ( #39706 ) chore(openai): update snapshots ( #39657 ) fix(openai): support o
At its core this is about langchain-ai/langchain, worth a look if that's in your workflow. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
BTS members around the text "Look what you can do" and "Unlock BTS" in a Google search bar
Builders evaluating Try 4 new interactive BTS experiences inside should verify the source before changing a production default. For builders, this can change a default in the review, editor, or build workflow you touch every day.
On the left: A blue background with black text saying 'Google' and 'Ask a scientist about flood forecasting.' On the right: Two Googlers in a virtual interview about flood forecasting
Builders evaluating Ask a Scientist should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
Google’s Gemini app offers full-length practice SAT tests at no cost. Here’s how to take these SAT practice tests as you prep.
The thing to notice is Keep your SAT prep on track with practice; decide whether it changes your next build. For builders, a release like this shifts the latency, capability, and cost trade-offs of what to run next.
We're sharing stories from the first graduates of America's Workforce Academy, Meta's free training program that guarantees trade jobs for all graduates.
The thing to notice is America’s Workforce Academy; decide whether it changes your next build. For builders, this can change a default in the review, editor, or build workflow you touch every day.
OpenAI and CodeAI are partnering to help students build AI literacy, think critically about AI, and develop the skills to use and shape it responsibly.
This lands on Partnering with CodeAI to prepare the first AI - gauge whether it shifts what you already ship. For builders, this may affect how you choose, deploy, or govern AI tools this week.
ChatGPT for Teens helps teens learn, think critically, and use AI with confidence, with stronger built-in protections, healthy-use features, and additional controls for parents.
This lands on Introducing ChatGPT for Teens - gauge whether it shifts what you already ship. For builders, this can change a default in the review, editor, or build workflow you touch every day.
OpenAI is strengthening monitoring, alignment, and security for frontier AI models. See how new safeguards are guiding the pace of model development.
This lands on Pacing model development in an era of cyber-critical - gauge whether it shifts what you already ship. For teams giving agents real access, this surfaces a failure mode worth weighing before widening agent or tool permissions.
A high-altitude view of a commercial airplane flying above a layer of clouds, with a distinct white contrail trailing behind it.
Teams affected by Operation Blue Skies need to decide whether its documented change alters their current workflow. For builders, this may affect how you choose, deploy, or govern AI tools this week.
Hermes Agent v0.20.4 (v2026.8.18) Release Date: August 18, 2026 Patch release. This tag rolls up the ~74 PRs merged since v0.20.3 into a stable tagged release for downstream consumers (Docker images, hosted deployments, fresh installs).
This lands on NousResearch/hermes-agent - gauge whether it shifts what you already ship. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
Asana used OpenAI Codex to replace an outdated testing system in two weeks, completing work expected to take five years for about $12K.
Teams affected by Asana cleared 5 years of engineering work need to decide whether its documented change alters their current workflow. For builders, this may affect how you choose, deploy, or govern AI tools this week.
arXiv:2607.14345v4 Announce Type: replace-cross Abstract: People use language models for practical questions whose answers are difficult to verify. We show that models exhibit covert value leakage: the information they provide is influenced by their own values, without this influence being disclosed to the user.
At its core this is about Value Leakage, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2607.19942v3 Announce Type: replace-cross Abstract: This work introduces G-MAD, an open-source framework that uses Arma3 to generate synchronized multi-view RGB-T data for aerial object detection. G-MAD addresses key limitations of real-world aerial dataset construction, including limited viewpoint control, imper
Teams affected by G-MAD need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.05863v2 Announce Type: replace Abstract: Modern models no longer keep a plain KV cache: latent caches, learned sparse selectors and recurrent states each carry the model's memory in a different form, and each fails differently under compression. We give a runtime observability contract that covers all four m
Operators using related systems should check whether Runtime Observability for Heterogeneous Attention Memory changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or...
arXiv:2607.28699v3 Announce Type: replace-cross Abstract: KV-cache quantization is validated today by offline benchmark averages; a deployed system cannot tell whether compression is damaging the request it is serving right now. We give it a provably sound runtime meter -- a "DTrace for KV quantization": a per-(layer
The thing to notice is WitCert; decide whether it changes your next build. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.10875v2 Announce Type: replace-cross Abstract: Large language model (LLM) agents are increasingly deployed as personal assistants. Existing evaluations, however, mostly use short, self-contained requests in static environments.
The thing to notice is VibeLifeBench; decide whether it changes your next build. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.13606v2 Announce Type: replace Abstract: The next generation of AI agents is increasingly moving beyond systems that answer isolated questions toward persistent personal assistants that can understand, remember, and continuously learn from users' experiences. Such assistants require long-term memory to accum
Operators using related systems should check whether MobileMem changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14552v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly evaluated and used in medicine, but clinical usefulness depends on answer accuracy and whether confidence tracks evidence quality and uncertainty. We developed a controlled, psychophysics-inspired clinical benchmark to test di
At its core this is about Large Language Models Show Metacognitive Sensitivity in Medical, worth a look if that's in your workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.14569v1 Announce Type: new Abstract: Neural solvers for constraint satisfaction problems have achieved remarkable in-distribution accuracy, yet they suffer from a fundamental limitation persistent constraint violations occur under distribution shifts even when the model reports high confidence. This position
Teams affected by Position need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14580v1 Announce Type: new Abstract: OGX (Open GenAI Stack) is an open-source AI application server and Python library that implements the APIs of major frontier labs (OpenAI, Anthropic, Google) with pluggable backend providers. Developers building agentic AI applications--such as retrieval-augmented generat
Operators using related systems should check whether OGX changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14613v1 Announce Type: new Abstract: Modern LLM-agent frameworks increasingly interoperate through standards such as Anthropic's Model Context Protocol (MCP) for agent-to-tool access and Google's Agent2Agent (A2A) protocol for agent delegation and negotiation. However, these protocols specify transport and d
At its core this is about Do LLM Agents Negotiate Rationally? A Mechanism-Design Framework, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14992v1 Announce Type: new Abstract: Language-model systems increasingly read from stores they also write to, so a claim that was merely written earlier can return looking retrieved. We tested whether the message package carrying an unsupported assignment changes which answer a model gives in a synthetic loo
What's actually new here is Does a Tool Result Carry More Authority Than - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15082v1 Announce Type: new Abstract: Cold chain logistics has advanced technologically, yet most deployed systems remain reactive monitors, not decision-making agents: thresholds trigger alerts, but nothing relates violations to cumulative product degradation or converts degradation signals into logistics de
What's actually new here is Beyond Thresholds - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15117v1 Announce Type: new Abstract: Analytical models of peak VRAM consumption for LLM inference decompose memory into weight-storage, KV-cache, and activation terms parameterized by step count, tool invocations, and context expansion. We evaluate this decomposition empirically within a strictly scoped meas
The thing to notice is Anatomy of a Quantized Agent; decide whether it changes your next build. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15354v1 Announce Type: new Abstract: LLMs are increasingly used in morally sensitive contexts, yet it is unclear whether they apply ethical principles consistently across situations. A model that can state a moral principle may still violate it when the same scenario is rephrased or reframed.
At its core this is about Incoherent by Design? On the Moral Self-Consistency, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15400v1 Announce Type: new Abstract: Large Language Models (LLMs) are notorious for struggling with assessing their own uncertainty, detecting knowledge conflicts, or recognizing when problems exceed their expertise; such limitations inevitably undermine reliability and trust in LLMs. In this paper, we prese
At its core this is about Implementation of a Metacognition Framework for Self-Awareness, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15703v1 Announce Type: new Abstract: Large language model (LLM) agents often perform poorly on complex, long-horizon tasks because their context becomes increasingly cluttered over time. As interactions accumulate, detailed execution traces and intermediate outputs dominate the context, making it difficult f
What's actually new here is HyMem - see if it moves anything you maintain. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.15746v1 Announce Type: new Abstract: We present a forensic analysis of the generation pipeline behind a recent AI-driven influence campaign. We introduce PROPAGIA, a corpus of 2,646 propagandist French articles from the Storm-1516/CopyCop campaign disclosed by VIGINUM and INSIKT GROUP in 2025.
What's actually new here is Propaganda Forensics - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15772v1 Announce Type: new Abstract: When a language model refuses to answer a prompt, it is unclear whether the correct answer is erased from its internal representations, or merely suppressed at the output layer. We investigate this mechanism using a controlled withhold setting, which yields perfectly matc
This lands on Broken Symmetry in LLM Refusal - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15797v1 Announce Type: new Abstract: KV-cache eviction caps the memory cost of long reasoning traces but is inherently lossy because the model decodes from a partial view of its history. Under aggressive budgets, this not only lowers accuracy but can also cause runaway degeneration, where the model produces
Operators using related systems should check whether KV-Rescue changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15893v1 Announce Type: new Abstract: The rise of social media bots poses a persistent threat, enabling misinformation, opinion manipulation, and the erosion of trust in online platforms. To combat this, machine learning systems have been developed to detect and limit bot activity, but attackers continuously
Operators using related systems should check whether Breaking and Defending LLM-Powered Social Media Bot Detection changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model...
arXiv:2608.15958v1 Announce Type: new Abstract: Deciding whether a Sokoban puzzle is solvable is PSPACE-complete (Culberson, 1997): solutions can be exponentially long and there is no short certificate to check. Solvability is also a fragile property, since even a single misplaced wall can silently render an entire puz
Teams affected by Solvable Sokoban Without a Solver via Diffusion need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16394v1 Announce Type: new Abstract: Generating regulation-compliant test scenarios is essential for validating safety-critical automotive systems, yet Large Language Models (LLMs) struggle to ground outputs in long, hierarchical standards. We present RegulaRAG, a Retrieval-Augmented Generation (RAG) pipelin
The thing to notice is Think Inside the Chunk; decide whether it changes your next build. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16435v1 Announce Type: new Abstract: In collection operations, accumulating payload progressively slows the vehicle, imposing a cumulative penalty on routing efficiency. An onboard drone can offset this penalty by retrieving outlying items, thereby shortening the makespan and increasing operational profit.
What's actually new here is Drive, Pack, Fly - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16628v1 Announce Type: new Abstract: Modern Multimodal Retrieval-Augmented Generation (M-RAG) systems are fundamentally limited by the binary connectivity paradigm of traditional simple graphs, which fails to capture the intricate, high-order correlations among heterogeneous entities, such as the N-ary relat
Builders evaluating Hypergraph-based Multimodal Retrieval-Augmented Generation with Incremental Refinement should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model...
arXiv:2608.16763v1 Announce Type: new Abstract: Financial document validation in production, such as payroll auditing, tax compliance, and loan underwriting, demands exceptional accuracy, consistency, and reproducibility under strict enterprise constraints. In practice, documents arrive with heterogeneous layouts and f
At its core this is about LAVA, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14606v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as synthetic survey respondents, but existing evaluations ask whether answers look plausible at the individual level. We argue the right question is psychometric: do LLMs preserve the joint distribution, latent structur
At its core this is about Plausible but Not Valid, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14646v1 Announce Type: cross Abstract: Interpretable representation learning remains a key challenge in modern neural computation, particularly when models are expected not only to perform but also to explain their reasoning. This paper introduces iFuzz-Meta, an interpretable fuzzy rule-based learning framew
This lands on iFuzz-Meta - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14710v1 Announce Type: cross Abstract: Predicting spatial gene expression from hematoxylin and eosin (H\&E)-stained images offers a cost-effective alternative to spatial transcriptomics (ST). However, existing methods treat H\&E images as generic visual inputs and ignore their intrinsic biological hierarchy
Teams affected by Path2ST need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14720v1 Announce Type: cross Abstract: Following the molecular discovery and synthesis revolutions, scalable automated structure elucidation from routine spectroscopic data remains an outstanding challenge. Despite decades of computational efforts, no existing system achieved reliable reasoning over unseen s
What's actually new here is Multi-Agent Closed-Loop Reasoning for Organic Structure Elucidation - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14737v1 Announce Type: cross Abstract: This study analyses LLMs in imbalanced binary classification, using study screening in systematic reviews as the application domain. An experiment was conducted in five reviews, comparing individual and batch processing, with and without prevalence metadata.
Teams affected by Class Imbalance and Batch Effects in LLM-Based Screening need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14763v1 Announce Type: cross Abstract: Assessment of ventriculomegaly (VM) on fetal brain ultrasound relies primarily on measuring lateral ventricular atrial width on standard planes, which is operator-dependent and may not fully reflect the overall ventricular enlargement. Fetal brain MRI provides more reli
The thing to notice is Cross-Modal Ultrasound-MRI Learning for Fetal Brain Ventricular Volumetry; decide whether it changes your next build. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14922v1 Announce Type: cross Abstract: Mechanistic interpretability has recently expanded to Vision Transformers (ViTs), with Sparse Autoencoders (SAEs) increasingly used as post-hoc tools to decompose internal representations into sparse and more interpretable features. However, because post-hoc SAEs are tr
This lands on SpIn-ViT - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15004v1 Announce Type: cross Abstract: Lung cancer remains one of the leading causes of cancer-related mortality worldwide, and Computed Tomography (CT) is a primary imaging tool for screening and followup assessment. After pulmonary nodule detection, radiologists manually assess anatomical location, diamete
At its core this is about FZ-VLM, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15006v1 Announce Type: cross Abstract: Although visual reasoning is crucial for solving complex geometry tasks, existing vision-language models rely heavily on text-only reasoning. Some recent methods introduce intermediate visual states to facilitate reasoning, but they are often hindered by inaccurate geom
Builders evaluating MetaReason should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15410v1 Announce Type: cross Abstract: Reasoning segmentation enables vision-language models (VLMs) to translate mission-relevant language requests into pixel-level visual grounding, offering a natural perception interface for embodied agents. However, existing benchmarks largely focus on generic visual scen
Builders evaluating FloodReasonBench should verify the source before changing a production default. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.15424v1 Announce Type: cross Abstract: The rapid adoption of large language models has enabled the development of clinical multi-agent systems (MAS) capable of integrating multimodal patient data and supporting increasingly complex clinical decision-making. However, the deployment of these systems in real-wo
Operators using related systems should check whether ETHOS changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15425v1 Announce Type: cross Abstract: Vision-language models (VLMs) achieve strong performance on high-level multimodal tasks, yet numerosity perception, a cognitive ability that emerges in human infants before language acquisition, remains poorly understood in current models, as existing counting benchmark
Builders evaluating NumerosityVLM should verify the source before changing a production default. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.15574v1 Announce Type: cross Abstract: Video question answering systems built on vision-language models often produce timestamped claims with high confidence even when unsupported by the cited frame. This deceptive hallucination arises because timestamps imply grounding without ensuring correctness, increasi
Builders evaluating Catching Hallucinated Citations in Video-LLM Question Answering should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15654v1 Announce Type: cross Abstract: Large language models can write fluent stories, but open-ended storytelling requires more than local fluency. In evolving world simulations and AI-native games, models must preserve facts, relationships, causal dependencies, and character states as the world changes.
At its core this is about When Stories Evolve, worth a look if that's in your workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.15731v1 Announce Type: cross Abstract: Deep Neural Networks (DNNs) deployed in high-risk domains, such as healthcare and autonomous driving, must be not only accurate but also understandable to ensure user trust. In real-world computer vision tasks, these models often operate on complex images containing bac
The thing to notice is Identifying Confusion Trends in Concept-based XAI for Multi-Label; decide whether it changes your next build. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15780v1 Announce Type: cross Abstract: Stale recommendations are a pervasive challenge and a leading source of user complaints on large-scale content platforms. Items lose relevance through two primary mechanisms: supersession, where emerging updates render prior coverage stale, and relevance decay, where an
This lands on Decomposing Staleness in Recommender Systems - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.15787v1 Announce Type: cross Abstract: Two Mixture-of-Experts (MoE) forward passes can share every weight yet route the same token through different experts. This creates a possible blind spot in same-weight self-distillation, where a demonstration-conditioned teacher supervises a query-only student.
Teams affected by Routing Divergence Is Not Evidence of Behavioral Influence need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent...
arXiv:2608.15828v1 Announce Type: cross Abstract: Current evaluation of metaphor explanations relies mainly on holistic quality ratings, revealing little about how explanation quality is structured or where human judgments agree and diverge. We introduce a cognitively motivated framework that decomposes metaphor explan
Teams affected by A Cognitively Motivated Multidimensional Framework for Evaluating Metaphor need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model...
arXiv:2608.16070v1 Announce Type: cross Abstract: Reliable global ocean forecasting is critical for climate monitoring, marine navigation, and extreme event early warning. Physics-based ocean forecasting models impose prohibitive computational costs, while existing deep learning approaches predominantly rely on structu
What's actually new here is OceanLight - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16181v1 Announce Type: cross Abstract: Recent advances in large language models have enabled a new class of agentic data science systems that allow users to complete complex data science workflows through natural language. Although these systems can significantly reduce manual effort, it remains difficult to
Teams affected by MUSE need to decide whether its documented change alters their current workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.16357v1 Announce Type: cross Abstract: Autonomous agents share a transport and can call each other's tools, but they cannot share what they know: no protocol lets two agents' memories reconcile a fact phrased two ways, link related facts held apart, or reconcile contradictory knowledge without silently disca
This lands on MELD - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.16539v1 Announce Type: cross Abstract: Large Audio Language Models (LALMs) have made rapid progress on standardized benchmarks, yet their deployment in practical media workflows, curation, archival indexing, and content distribution remains largely unrealized. We identify automated audio chapterization, the
What's actually new here is Listen, Reason, and Segment - see if it moves anything you maintain. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.16837v1 Announce Type: cross Abstract: Humanoid robots hold great promise as general-purpose agents in human-centered environments, yet generalist vision-language-action (VLA) foundation models are not readily applicable to humanoid whole-body loco-manipulation. The high dimensionality and interdependence of
At its core this is about HAF, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2509.15035v2 Announce Type: replace Abstract: This study investigates the use of generative AI to support formative assessment through machine generated reviews of peer reviews in graduate online courses in a public university in the United States. Drawing on Systemic Functional Linguistics and Appraisal Theory
What's actually new here is Calibrated Generative AI as Meta-Reviewer - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2510.10813v2 Announce Type: replace Abstract: Large Language Models (LLMs) are increasingly applied to domains that require reasoning about other agents' behavior, such as negotiation, policy design, and market simulation. However, can we trust LLMs to think strategically in complex situations?
Operators using related systems should check whether The Fragility of Strategic Thinking in Large Language changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or...
arXiv:2511.17006v2 Announce Type: replace Abstract: Scaling test-time computation has been extended from language model reasoning to tool-augmented agents, where scaling involves not only thinking in tokens but also acting via tool calls that directly constrain environmental interaction. However, we found that simply i
Teams affected by Budget-Aware Tool Use Enables Effective Agent Scaling need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2606.03988v3 Announce Type: replace Abstract: Vision language models (VLMs) excel at many tasks but still struggle with spatial reasoning when critical information is not directly observable. Many such problems require imaginative perception: inferring what would be seen from an unseen viewpoint, tracing paths th
Builders evaluating Imaginative Perception Tokens Enhance Spatial Reasoning in Multimodal should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent...
arXiv:2606.14582v2 Announce Type: replace Abstract: Efficient route optimization play a vital role in ensuring both safety and punctuality in railway operations. It is very crucial particularly in heterogeneous multi-gauge railway networks with varying train speed, stopping pattern, infrastructure compatibility constra
The thing to notice is A Temporal Planning Framework for Disruption Aware Dynamic; decide whether it changes your next build. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2405.07780v3 Announce Type: replace-cross Abstract: This paper explores test-agnostic long-tail recognition, a challenging long-tail task where the test label distributions are unknown and arbitrarily imbalanced. We argue that the variation in these distributions can be broken down hierarchically into global and
Operators using related systems should check whether DirMixE changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2501.12147v2 Announce Type: replace-cross Abstract: Selecting appropriate training data is crucial for instruction fine-tuning of large language models (LLMs), which aims to (1) elicit strong capabilities, and (2) achieve balanced performance across different tasks. Influence-based methods show promise in achievi
Operators using related systems should check whether Improving Influence-based Instruction Tuning Data Selection for Balanced changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation...
arXiv:2506.07449v2 Announce Type: replace-cross Abstract: Recent advances in Large Language Models (LLMs) have driven their adoption in recommender systems through Retrieval-Augmented Generation (RAG) frameworks. However, existing RAG approaches predominantly rely on flat, similarity-based retrieval that fails to lever
Teams affected by LlamaRec-LKG-RAG need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2508.07195v2 Announce Type: replace-cross Abstract: Recent advances have demonstrated that Large Language Models (LLMs) can be effectively adapted for time series forecasting, revealing strong potential beyond natural language tasks. However, their performance remains constrained by two fundamental challenges: th
Teams affected by Adapting LLMs to Time Series Forecasting via Temporal need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2511.23043v2 Announce Type: replace-cross Abstract: We present a probabilistic data-driven weather model providing ensembles of high spatial resolution realizations of 87 variables at arbitrary ensemble size and forecast length. The model uses a global stretched grid, dedicating 2.5 km resolution to our Nordic re
Builders evaluating High-Resolution Probabilistic Data-Driven Weather Modeling with a Stretched-Grid should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model choice, or...
arXiv:2512.04032v4 Announce Type: replace-cross Abstract: We present jina-vlm, a token-efficient 2.4B parameter vision-language model that achieves state-of-the-art multilingual VQA performance among open 2B-scale VLMs. The model couples a SigLIP2 vision encoder with a Qwen3 language decoder and makes use of image tili
Operators using related systems should check whether jina-vlm changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2601.11496v3 Announce Type: replace-cross Abstract: AI agents increasingly mediate bargaining, negotiation and persuasion for people and firms. Such markets extend software-mediated commerce, but add a governance problem: independent model releases change delegates available to participants.
This lands on Sequential LLM Release Facilitates Manipulation in Regulated Markets - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2602.21819v3 Announce Type: replace-cross Abstract: Reconstructing dynamic visual experiences from brain activity provides a compelling avenue for exploring the neural mechanisms of human visual perception. While recent progress in fMRI-based image reconstruction has been notable, extending this success to video
Operators using related systems should check whether SemVideo changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2603.17555v2 Announce Type: replace-cross Abstract: Diffusion-based image-to-video (I2V) models are increasingly effective, yet they struggle to scale to ultra-high-resolution inputs (e.g., 4K). Generating videos at the model's native resolution often loses fine-grained structure, whereas high-resolution tiled de
Teams affected by FrescoDiffusion need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2604.09860v4 Announce Type: replace-cross Abstract: The pursuit of general-purpose robotics has yielded impressive foundation models, yet simulation-based benchmarking remains a bottleneck due to rapid performance saturation and a lack of true generalization testing. Existing benchmarks often exhibit significant
This lands on RoboLab - gauge whether it shifts what you already ship. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2605.27599v3 Announce Type: replace-cross Abstract: Agentic AI workloads - where a single user goal triggers multi-step orchestration, tool calls, retries, and failure recovery - are being targeted for edge deployment, with NVIDIA, Dell, HP, ASUS, MSI, Acer, and Gigabyte all shipping GB10-based desktop AI systems
This lands on The Energy Blind Spot - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.00576v2 Announce Type: replace-cross Abstract: High-polyphony symbolic music is increasingly used in generation, analysis, and arrangement, yet many downstream tasks require bounded representations with fixed tracks or slots. Converting richly orchestrated scores into compact forms is therefore necessary, bu
Teams affected by UOT-IR need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.11657v2 Announce Type: replace-cross Abstract: We introduce Semantic Lenia, an artificial life framework that transforms Large Language Model (LLM) inference from a static optimization problem into a continuous, closed-loop dynamical system. By establishing a non-linear homeostatic feedback loop to dynamical
Builders evaluating Semantic Lenia should verify the source before changing a production default. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
Initial release release(openai): 1.5.2a1 ( #39709 ) feat(openai): extract gateway metadata from response headers when available ( #39706 ) chore(openai): update snapshots ( #39657 ) fix(openai): support o-series models in get_num_tokens_from_messages ( #38710 ) release(openai): 1.5.1 ( #39653 ) fix(openai): preserve st
Teams affected by langchain-ai/langchain need to decide whether its documented change alters their current workflow. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
Multi-Vector (Late Interaction) Embedding Models with Sentence Transformers
Teams affected by Multi-Vector need to decide whether its documented change alters their current workflow. For builders, a release like this shifts the latency, capability, and cost trade-offs of what to run next.
Changes since langchain-core==1.5.5 chore(core): release 1.5.6 ( #39704 ) feat(core): incorporate gateway metadata to traces ( #39703 )
The thing to notice is langchain-ai/langchain; decide whether it changes your next build. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
What's changed Added the optional CLAUDE_CODE_PROJECT_DIR_NAME environment variable: hosts that give each session its own config directory can choose a short name for the per-project transcript directory Added the selection:clear keybinding action, so a key can be bound to clear an in-app text selection; also works in
Builders evaluating anthropics/claude-code should verify the source before changing a production default. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
Same Cluster, 33 Points More Utilization: What Changed Was the Order
Builders evaluating Same Cluster, 33 Points More Utilization should verify the source before changing a production default. For builders, this may affect how you choose, deploy, or govern AI tools this week.
Hermes Agent v0.20.3 (v2026.8.16.2) Release Date: August 16, 2026 Patch release. This tag rolls up the ~125 PRs merged since v0.20.2 into a stable tagged release for downstream consumers (Docker images, hosted deployments, fresh installs).
Teams affected by NousResearch/hermes-agent need to decide whether its documented change alters their current workflow. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
Low-angle view of a soccer player kicking a ball mid-air against a bright blue sky, with grass flying from their cleats.
What's actually new here is Get closer to the game with Gemini - see if it moves anything you maintain. For builders, a release like this shifts the latency, capability, and cost trade-offs of what to run next.
AI is reshaping cybersecurity for attackers and defenders alike. Learn how OpenAI is strengthening its defenses and what security teams can do now.
Operators using related systems should check whether The Defender’s Window changes compatibility, cost, or access requirements. For teams giving agents real access, this surfaces a failure mode worth weighing before widening agent or tool permissions.
OpenAI joins PORTS-Pike project, expanding community investment and supporting thousands of Southern Ohio jobs
Teams affected by OpenAI joins PORTS-Pike project need to decide whether its documented change alters their current workflow. For builders, this may affect how you choose, deploy, or govern AI tools this week.
arXiv:2605.01189v3 Announce Type: replace Abstract: Clinical AI adoption is hindered by the black-box/grey-box nature of high-performing models, which lack the ontological grounding and narrative transparency required for professional-level explainability. We present NEURON, a neuro-symbolic system designed to enhance
What's actually new here is NEURON - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2607.29516v2 Announce Type: replace-cross Abstract: AI coding agents are generating code at volumes that exceed the capacity of traditional peer review. At the same time, existing AI code review tools over-index on low-value suggestions such as style and best practices while under-indexing on the concerns human r
This lands on From Code Review to Code Critique - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.13987v1 Announce Type: new Abstract: Nanbeige4.2-3B is a 3B-parameter agentic model built around a Looped Transformer (LT) that reuses one stack of layers for a second forward pass, adding effective depth without additional parameters. Evaluated on Apple Silicon (MPS), we identify five independent bugs which
The thing to notice is Nanbeige4.2-3B on Apple Silicon; decide whether it changes your next build. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14246v1 Announce Type: new Abstract: In today's fast-paced environment, the ability to swiftly access, understand, and act on data is no longer optional; it is essential. Yet most organizations remain data-rich but insight-poor, constrained by the complexity of querying, interpreting, and explaining enterpri
What's actually new here is Polaris - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14252v1 Announce Type: new Abstract: Recent work suggests that some large language model representations have content or reference. Grounding can secure either without supplying live routes for correction.
This lands on Grounding Without Corrective Control - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.13563v1 Announce Type: cross Abstract: Early-stage teams often lack users, time, and budget to run repeated UX studies, yet still need decision-oriented signals to iterate safely. We study an LLM-driven UX micro-simulation pipeline that generates structured customer-experience feedback (walkthrough steps, fr
This lands on Proxy-Validated LLM UX Micro-Simulations - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.13572v1 Announce Type: cross Abstract: We present The Architect, a system that turns Microsoft Excel into an interactive view of deep learning mathematics. A user describes a neural network in a compact table.
What's actually new here is The Architect - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.13580v1 Announce Type: cross Abstract: Jais 2 is a family of Arabic-centric large language models developed jointly by MBZUAI, Cerebras, and Inception, designed to advance Arabic-centric language modeling, with strong performance across the Arabic and culturally grounded benchmarks evaluated in this report.
This lands on Jais 2 - gauge whether it shifts what you already ship. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.13675v1 Announce Type: cross Abstract: Between October 2018 and July 2026 AI models progressed from simple systems like BERT to massive agents that solve complex math and write software. The ability to resolve real coding issues improved by nearly six times per year since late 2024.
Operators using related systems should check whether From BERT to Frontier Agents changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.13681v1 Announce Type: cross Abstract: Translating C code into safe, idiomatic Rust is a longstanding software-engineering goal because it can eliminate entire classes of memory-safety vulnerabilities while preserving the functional behavior of legacy systems. Large language models (LLMs) have shown promise
Teams affected by Fine-Tuning Qwen3-27B for C-to-Rust Code Translation need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.13905v1 Announce Type: cross Abstract: HTTPS website fingerprinting (WF) aims to identify visited websites from metadata observable in encrypted traffic. However, real-world deployments introduce a significant out-of-distribution (OOD) problem caused by temporal and geographic changes, while previously unsee
At its core this is about CipherSight, worth a look if that's in your workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14016v1 Announce Type: cross Abstract: Live game commentary is scarce: it exists for professional esports broadcasts and almost nowhere else. We present a content-based video narration system that produces spoken, esports-style commentary for arbitrary gameplay recordings using a general-purpose vision-langu
Teams affected by Content Based Video Narration of Gameplay with Vision need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14058v1 Announce Type: cross Abstract: Seismic facies segmentation has emerged as a significant challenge in geophysics, requiring robust methods and systems to effectively identify geologically analogous facies with limited labeled data. Although existing studies have shown promising results in 2D facies se
Teams affected by Voxel-based 3D Facies Segmentation from Seismic Data need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.14172v1 Announce Type: cross Abstract: Text-to-image diffusion models have two major drawbacks that severely limit their practical utility: (1) standard models lack an intrinsic mechanism for continuous, concept-specific guidance (e.g., for precisely controlling how aesthetically pleasing an image looks), an
Teams affected by Concept Guidance need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2509.14704v3 Announce Type: replace Abstract: Benchmark saturation and training-data contamination increasingly obscure whether reported gains in large language models (LLMs) reflect genuine advances in reasoning or familiarity with recurring patterns in benchmark problems. We introduce the NazoNazo Benchmark, a
At its core this is about The Metacognitive Bottleneck, worth a look if that's in your workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2607.14975v2 Announce Type: replace Abstract: Channel foundation models (CFMs) are commonly evaluated in model-specific pipelines that differ in data, radio configurations, partitions, adaptation procedures, task definitions, and metrics, preventing reproducible comparison across CFMs and against task-specific ne
At its core this is about CFM-Bench, worth a look if that's in your workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.01856v2 Announce Type: replace Abstract: Bi-temporal remote-sensing disaster change captioning often needs to identify sparse and spatially localized changes across large pre- and post-event scenes and then translate them into coherent, factual descriptions. However, existing change captioning methods always
Teams affected by EchoChange need to decide whether its documented change alters their current workflow. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2608.02606v2 Announce Type: replace Abstract: Fault tolerance in classical computing has traditionally relied on static strategies like hardware redundancy and error-correcting codes. Biological systems, in contrast, exhibit adaptive plasticity, maintaining function through dynamic re-organisation around damage.
What's actually new here is Self-Organising Digital Circuits - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2601.22631v2 Announce Type: replace-cross Abstract: The application of data-driven remaining useful life (RUL) prediction has long been constrained by the availability of large amount of degradation data. Mainstream solutions such as domain adaptation and meta-learning still rely on large amounts of historical de
This lands on PEFT-MuTS - gauge whether it shifts what you already ship. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2605.25420v2 Announce Type: replace-cross Abstract: Large language model safety evaluation remains heavily English-centered, leaving low-resource languages under-measured even when models are deployed globally. We evaluate four open-weight instruction-tuned models on SomaliBench v0, a native-author-verified bench
What's actually new here is SomaliBench Eval - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent design.
arXiv:2606.08123v2 Announce Type: replace-cross Abstract: Model selection for safety-relevant visual recognition is often based on clean aggregate performance, although robustness, transfer, embedded latency, and explanation faithfulness may produce different preferences. This study presents a Human-Centered Benchmarki
Operators using related systems should check whether Beyond aggregate scores changes compatibility, cost, or access requirements. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.