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/.
What's changed Bug fixes and reliability improvements
Operators using related systems should check whether anthropics/claude-code changes compatibility, cost, or access requirements. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
This week’s updates across GitHub Copilot in the desktop app, CLI, and VS Code help you resume and organize work, review changes, and ask questions without losing context. GitHub Copilot...
What's actually new here is GitHub Copilot weekly releases - August 3 - see if it moves anything you maintain. For builders, this can change a default in the review, editor, or build workflow you touch every day.
Enterprise owners can now install public GitHub Apps created outside their enterprise on their enterprise account. This allows third-party integrators to build apps for enterprise management scenarios.
Teams affected by Enterprises can now install third-party GitHub Apps need to decide whether its documented change alters their current workflow. For enterprise teams, this moves the admin, budget, or governance controls needed to scale AI safely.
The Copilot impact dashboard now includes a “Potential return on investment” section that connects what you spend on Copilot to the pull request output you get back. What’s new Two...
This lands on Copilot impact dashboard adds a return on investment - 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.
Lite and Balanced effort levels for GitHub Copilot code review are now generally available. They let you match the depth of a review to the complexity and risk of a...
At its core this is about Copilot code review effort levels are generally available, 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.
Changes since checkpointpostgres==3.1.1 release(checkpoint-postgres): 3.1.2 ( #8565 ) release(checkpoint): 4.2.0 ( #8563 ) test(checkpoint-postgres,checkpoint-sqlite): run the conformance suite ( #8537 ) fix(checkpoint-postgres): find plain-value seeds when walking delta history ( #8535 ) chore: enable RUF100 and clear
Builders evaluating langchain-ai/langgraph 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.
Changes since checkpoint==4.1.1 release(checkpoint): 4.2.0 ( #8563 ) fix(checkpoint): collect writes at plain-value seed in delta channel history ( #8526 ) chore: enforce PLC0415 in tests for the remaining packages ( #8547 ) chore: enable RUF100 and clear unused noqa directives ( #8546 ) chore(deps): bump the minor-and
What's actually new here is langchain-ai/langgraph - see if it moves anything you maintain. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
Since agent apps arrived on GitHub, teams have been able to run agents from partners like Claude and Codex directly in their GitHub workflows. The Copilot usage metrics API now...
What's actually new here is Copilot usage metrics API adds agent app activity - see if it moves anything you maintain. For builders, this can change a default in the review, editor, or build workflow you touch every day.
TutorMoments: Do AI tutors know when to help and when to hold back?
Operators using related systems should check whether TutorMoments changes compatibility, cost, or access requirements. For builders, this may affect how you choose, deploy, or govern AI tools this week.
GitHub is making it easier to connect related work and manage multiple field values across issues and projects. Connect related issues The new “Relates to” issue relationship is available in...
Teams affected by “Relates to” issue relationship in public preview 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.
0.121.0 (2026-08-07) Full Changelog: v0.120.2...v0.121.0 Features api: add mid-conversation-tool-changes-2026-07-01 beta ( c7d1531 ) api: add support for session budgets, advisor tool, pinned inference location and skills auto-loading from GitHub ( 193bae0 ) Chores api: remove retired Claude Opus 4.1 models ( 5352a33 )
The thing to notice is anthropics/anthropic-sdk-python; decide whether it changes your next build. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
This release expands secret scanning’s coverage with more secrets blocked by push protection, one new secret scanning partner, and richer metadata on alerts. Secret scanning partnership program Lovable Labs is...
At its core this is about Secret scanning coverage updates, 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 is sharing preliminary cybersecurity evaluations for Astra and the steps we’re taking to strengthen safeguards and security controls.
Operators using related systems should check whether Responding to the next frontier of critical cyber changes compatibility, cost, or access requirements. For teams giving agents real access, this surfaces a failure mode worth weighing before widening agent...
Enabling GitHub Code Quality on a repository no longer creates a ruleset that automatically requests a code review from GitHub Copilot on your pull requests. In repositories that already have...
This lands on GitHub Code Quality no longer adds Copilot - 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.
Google AI published a source-backed AI update around See what 5 builders are making with Gemini Omni. RepoRadar is keeping the source link direct for verification.
Builders evaluating See what 5 builders are making with Gemini 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.
Changes since langchain-openai==1.4.1 release(openai): 1.4.2 ( #39322 ) fix(openai): handle ContextWindowExceededError ( #39300 ) chore: bump the minor-and-patch group across 3 directories with 7 updates ( #39187 ) fix(openai): filter langchain-generated content block IDs ( #39209 ) fix(openai): preserve Responses text
Operators using related systems should check whether langchain-ai/langchain changes compatibility, cost, or access requirements. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
Discover how HSP GRUPPE uses ChatGPT Enterprise to boost productivity, improve work quality, and create more capacity for tax advisory and client service.
The thing to notice is How HSP GRUPPE builds AI capabilities for tax; decide whether it changes your next build. For enterprise teams, this moves the admin, budget, or governance controls needed to scale AI safely.
What's changed Added self-hosted environments: claude self-hosted-runner turns your own machines or containers into a place Claude Code web, mobile, and desktop sessions can run, on Team and Enterprise plans Added archive plugin source: install plugins from a zip over HTTPS without git or npm, with optional SHA-256 pin
Teams affected by anthropics/claude-code 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.
arXiv:2608.05225v1 Announce Type: new Abstract: Research agents can increasingly search literature, propose hypotheses, generate code, run experiments, and draft manuscripts from a single topic. However, a research project is not merely a larger task: it is a long-horizon agenda that must be advanced through multiple b
Builders evaluating Project2Task should verify the source before changing a production default. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.05233v1 Announce Type: new Abstract: Dreams can be emotionally intense but difficult to communicate. We describe the Dream Scene Visualiser (DSV) system which turns written dream descriptions into a temporal sequence of four panel images visualising the dream.
Builders evaluating Coherence-Oriented Dream Scene Visualisation 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:2608.05248v1 Announce Type: new Abstract: Generating large-scale, freely explorable 3D worlds from open-ended text remains challenging because a system must jointly maintain global spatial coherence, rich local content, and explicit assets suitable for downstream editing and reuse. We present WorldClaw, a fully a
The thing to notice is WorldClaw; 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:2608.05367v1 Announce Type: new Abstract: Counterfactual analysis aims to predict potential outcomes under hypothetical scenarios, offering valuable insights for decision-making. This paper investigates the application of large language models (LLMs), specifically the GPT-3.5 model, for counterfactual analysis.
This lands on Counterfactual Analysis via Large Language Models - 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:2608.05893v1 Announce Type: new Abstract: Electrocardiography (ECG) is one of the most widely used non-invasive tools for diagnosing cardiovascular disease, but transforming multi-lead ECG recordings into reliable clinical reports remains challenging. Automating ECG report generation could reduce clinicians' inte
At its core this is about ECG-LENS, 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:2608.05948v1 Announce Type: new Abstract: Physics engines facilitate large-scale training and evaluation for embodied intelligence, while generative video world models are emerging as implicit simulators of future states and interactions. However, existing evaluations of physical fidelity are often conducted in i
Builders evaluating GAUGE should verify the source before changing a production default. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.06085v1 Announce Type: new Abstract: Survey-country metadata can improve an LLM's forecast of an individual response when informative, yet the same cue may redirect the forecast when assigned at random. A within-record audit tests whether disclosing a random label's uniform, record-independent origin reduces
The thing to notice is Signal or Spurious Cue? A Randomized Audit; 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:2608.06351v1 Announce Type: new Abstract: This paper addresses the limitations of Explainable Artificial Intelligence (XAI) with respect to insufficient evaluation. They are illustrated through the DetoxAI image recognition system for bias detection and concept unlearning.
The thing to notice is Challenges in Evaluating Explanation Methods for Static; 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:2608.05170v1 Announce Type: cross Abstract: Role-playing agents (RPAs) have emerged as a key application of large language models, enabling immersive and high-fidelity character simulation. Accurate role-playing of established characters requires not only stylistic imitation but also temporally consistent and cau
Builders evaluating DREAM 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:2608.05188v1 Announce Type: cross Abstract: Despite ever-increasing sophistication in language model (LM) pre- and post-training pipelines, many important failures persist: models overcondition on user framing ("sycophancy"), exhibit incomplete logical generalization, and produce confident but incorrect responses
At its core this is about Position, 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:2608.05210v1 Announce Type: cross Abstract: Picture books and comics have long been used to disseminate hateful narratives because they are easily understood even by children, as exemplified by the notorious Nazi propaganda picture book \emph{Der Giftpilz}. Recently, frontier text-to-image (T2I) systems such as G
Teams affected by Innocent Panels, Hateful Stories 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:2608.05255v1 Announce Type: cross Abstract: Retail investors lack access to the kind of personalized, tax-aware portfolio management that institutional clients take for granted -- existing robo-advisors use static, rule-based allocation, and institutional-grade systems require account minimums and technology stac
What's actually new here is An Emerging Retail Portfolio Management Application - 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:2608.05545v1 Announce Type: cross Abstract: Generative AI used as a capable servant has greatly accelerated intellectual work, but it also risks eroding human epistemic agency by encouraging uncritical acceptance of AI-generated reasoning. This creates a need for mechanisms that preserve human agency by augmentin
What's actually new here is Vibe Compiler - 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:2608.05806v1 Announce Type: cross Abstract: While standard Next-Token Prediction (NTP) lays the foundation of language model pre- training, its teacher-forced training paradigm may not be optimal for long-horizon reasoning and planning. Recent works such as Multi-Token Prediction (MTP) and Next-Latent prediction
At its core this is about Hierarchical Latent Prediction for Language Models, worth a look if that's in your workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.06015v1 Announce Type: cross Abstract: Deep Reinforcement Learning (RL) is notoriously sample inefficient. One contributing factor is that RL agents are typically initialized from scratch, forcing them to acquire task-relevant knowledge through online interaction.
Operators using related systems should check whether ProDVI 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:2608.06105v1 Announce Type: cross Abstract: Artificial Intelligence (AI)-assisted navigation can help Arctic shipping adapt to rapidly changing sea-ice conditions, but reliable deployment requires reward models that are interpretable and robust to changing environments. Inverse reinforcement learning (IRL) provid
Teams affected by Does Latent Context Help? A Controlled Evaluation 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:2608.06202v1 Announce Type: cross Abstract: Large language model (LLM) benchmark evaluations are routinely used to support claims about model safety, reliability, and deployment readiness. Yet most evaluations rely on a single access modality (model APIs), perform a single run per prompt, and report accuracy as t
Builders evaluating What Current AI Benchmarks Leave Unmeasured should verify the source before changing a production default. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2608.06236v1 Announce Type: cross Abstract: Our primary objective is to advance video object counting in crowded scenes, aiming to robustly count all instances of a target category based on given text or visual prompts. Existing methods rely on RGB information, limiting their discriminative ability in crowded and
The thing to notice is Depth-Guided Video Object Counting in Crowded Scenes; 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:2509.26331v2 Announce Type: replace Abstract: The rapid advancement of LLMs sparked significant interest in their potential to augment or automate managerial functions. One of the most recent trends in AI benchmarking is performance of Large Language Models (LLMs) over longer time horizons.
Teams affected by AI Playing Business Games 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:2601.21800v4 Announce Type: replace Abstract: We introduce BioAgent Bench, an evaluation suite designed for measuring the performance and robustness of AI agents in common bioinformatics tasks. The suite consists of manually curated end-to-end tasks (e.g., RNA-seq, variant calling, metagenomics) accompanied by ta
This lands on BioAgent Bench - 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:2603.29152v3 Announce Type: replace Abstract: Metal-organic frameworks (MOFs) offer a vast design space, and as such, computational simulations play a critical role in predicting their structural and physicochemical properties. However, MOF simulations remain difficult to access because reliable analysis require
At its core this is about SimMOF, 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:2604.02721v3 Announce Type: replace Abstract: Competitive programming remains one of the last few human strongholds in coding against AI. The best AI system to date still underperforms the best humans competitive programming: the most recent best result, Google's Gemini~3 Deep Think, attained 8th place even not b
The thing to notice is GrandCode; 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:2605.00737v3 Announce Type: replace Abstract: Agentic AI architectures augment LLMs with external tools, unlocking strong capabilities but potentially incurring substantial costs. Moreover, tool use is not always beneficial: redundant or low-utility calls can even harm task performance.
What's actually new here is To Call or Not to Call - 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:2605.25815v4 Announce Type: replace Abstract: Agent-to-Agent (A2A) networks enable autonomous AI agents to collaborate by sharing reusable problem-solving instructions. However, how these decentralized ecosystems operate in practice remains largely unexplored.
Builders evaluating Behind EvoMap 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:2606.31134v3 Announce Type: replace Abstract: While Large Language Models (LLMs) have demonstrated exceptional capabilities in mathematical reasoning, they frequently produce subtle errors that evade human detection. Formal mathematical languages like Lean 4 offer mechanical proof checking, strongly motivating th
This lands on Beyond the Library - 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:2502.14671v4 Announce Type: replace-cross Abstract: Large Language Model (LLM) representations are known to align with brain activity during language processing, but it remains unclear what drives this alignment. We test whether explainable AI (XAI) can help answer this: using attribution methods, we quantify the
This lands on Explanations of Large Language Models Explain Language Representations - 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:2512.02551v4 Announce Type: replace-cross Abstract: In this paper, we propose CUDA-L2, a system that combines large language models (LLMs) and reinforcement learning (RL) to automatically optimize Half-precision General Matrix Multiply (HGEMM) CUDA kernels. Using CUDA execution speed as the RL reward, CUDA-L2 aut
Builders evaluating CUDA-L2 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:2604.01170v2 Announce Type: replace-cross Abstract: While test-time scaling has enabled large language models to solve highly difficult tasks, state-of-the-art results come at exorbitant compute costs. These inefficiencies can be attributed to the miscalibration of post-trained language models, and the lack of ca
What's actually new here is Online Reasoning Calibration - 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:2604.17359v2 Announce Type: replace-cross Abstract: Language models asked to simulate psychiatric patients produce cases that survive inspection one at a time and populations that match no real one. We gave GPT-4o-mini, Gemini-3-Flash, DeepSeek-V3 and GLM-4.7 each of 120 demographic cohorts under two framings, on
What's actually new here is Plausible Patients, Impossible Populations - 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:2608.02616v2 Announce Type: replace-cross Abstract: We present what is, to our knowledge, the first systematic evaluation of OpenAI's Privacy Filter (OPF), a 1.5B-parameter model that converts an autoregressive language model into a bidirectional PII detector, across 32 benchmarks spanning 14 languages and 5 doma
At its core this is about OpenAI Privacy Filter, worth a look if that's in your workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
Enterprise owners can now centrally control which Model Context Protocol (MCP) servers GitHub Copilot clients are allowed to run by using the new allowedMcpServers and deniedMcpServers keys in enterprise managed...
Teams affected by MCP allowlists in enterprise managed settings need to decide whether its documented change alters their current workflow. For builders, this can change a default in the review, editor, or build workflow you touch every day.
You can now set pull request limits at the organization level. Pull request limits let you define how many open pull requests a user without write access can have at...
Builders evaluating Set pull request limits at the organization level 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.
What's changed [VSCode] Added Focus view: a chat-menu toggle that hides tool activity behind an expandable per-turn summary with a live running-tool indicator, toggled with Ctrl+Alt+F or the "Claude Code: Toggle Focus view" command Added mode: "mask" for sandbox credential files on Linux and WSL - sandboxed commands re
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.
Enterprise administrators can now customize managed settings by targeting enterprise teams with itemized configuration files. Large enterprises can scale governance without bottlenecking every configuration change through central administrators or one-size-fits-all...
This lands on Enterprise team specialization for managed settings - gauge whether it shifts what you already ship. For enterprise teams, this moves the admin, budget, or governance controls needed to scale AI safely.
If your repository restricts issue creation to collaborators only, users with the triage role can now create issues. Previously, the restricted setting only allowed users with write access to create...
Operators using related systems should check whether Triage role can bypass issue creation restrictions changes compatibility, cost, or access requirements. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
Hermes Agent v0.20.0 (v2026.8.3) Release Date: August 3, 2026 Since v0.19.0: ~3,650 commits · ~1,400 merged PRs · ~5,200 files changed · ~559,000 insertions · ~405,000 deletions · ~1,200 issues closed · 650+ contributors The Herald Release. Hermes is the herald of the gods, and this release makes him one in earnest: he
At its core this is about NousResearch/hermes-agent, 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.
Migrations from gitlab.com and GitLab Self-Managed to GitHub Enterprise Cloud with GitHub Enterprise Importer (GEI) are now generally available. You can now self-serve these migrations using GEI and the gh...
Builders evaluating Migrate from GitLab to GitHub with GitHub Enterprise should verify the source before changing a production default. For enterprise teams, this moves the admin, budget, or governance controls needed to scale AI safely.
Google AI published a source-backed AI update around Welcome to Sail Tower, our newest Austin office. RepoRadar is keeping the source link direct for verification.
This lands on Welcome to Sail Tower, our newest Austin office - gauge whether it shifts what you already ship. For builders, this may affect how you choose, deploy, or govern AI tools this week.
Illustrations of a laptop, an AI spark, messages, code, and a 3-D cube
Builders evaluating Inside our 353,000-person vibe coding course should verify the source before changing a production default. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
GPT-Live enables continuous voice interaction with AI, using a turnless speech model and low-latency architecture for faster, more natural conversations.
This lands on How we built a realtime system for responsive - gauge whether it shifts what you already ship. For builders, a release like this shifts the latency, capability, and cost trade-offs of what to run next.
arXiv:2603.00910v3 Announce Type: replace-cross Abstract: Layer-wise capacity in large language models is highly non-uniform: some layers contribute disproportionately to loss reduction, whereas others are nearly redundant. Existing layer-scoring methods provide sensitivity estimates but do not give a principled rule f
This lands on Curvature-Weighted Capacity Allocation - 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:2607.14144v2 Announce Type: replace Abstract: The Platonic Representation Hypothesis (PRH) holds that as models scale, representations of heterogeneous networks converge toward a shared model of reality. We propose its sequel and boundary, the Capability Convergence Hypothesis (CCH): under a fixed per-token infer
The thing to notice is The Capability Convergence Hypothesis; 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:2607.23886v3 Announce Type: replace-cross Abstract: X-ray absorption spectroscopy (XAS) is central to understanding the local electronic and atomic structure of materials, yet most published spectra remain inaccessible to data-driven analysis because they are embedded in figures and described through fragmented t
What's actually new here is Harnessing X-ray Absorption Spectroscopy Data through Multimodal Mining - 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:2607.28662v1 Announce Type: new Abstract: Large language models extract entities and relationships from unstructured documents fluently but inconsistently: type vocabularies fracture across documents, the same person surfaces under several name variants, relationships duplicate, and distinct individuals who share
Teams affected by An Ontology-Guided, Deduplication-Aware Extraction Layer for Knowledge Graph 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:2607.28956v1 Announce Type: new Abstract: Large language model agents are increasingly evaluated as autonomous tool users, yet most benchmarks focus on bounded tasks with immediate success criteria. Real-world deployments often require Long-Term Coherence, the capacity to preserve purposeful behavior across exten
What's actually new here is MerchantBench - see if it moves anything you maintain. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2607.29077v1 Announce Type: new Abstract: Counterfactual explanations (CEs) enhance the interpretability of machine learning models by identifying the smallest change to an input required to obtain a desired output. Although CEs are conventionally formulated as a distance-minimization problem, the theoretical bas
Teams affected by A Generalized-Bayes Perspective on Counterfactual Explanations 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:2607.29440v1 Announce Type: new Abstract: Long-term multimodal memory must support not only retrieving relevant information but also computing over observations accumulated across interactions. Existing systems largely emphasize \emph{retrieval memory}, organizing interaction histories through summaries and index
The thing to notice is Beyond Retrieval; 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:2607.29677v1 Announce Type: new Abstract: Enterprise workflows increasingly rely on agents for \emph{schema-guided extraction}: given a document and a user-defined schema, the agent faithfully follows the schema to produce the correct output with source evidence as grounding metadata. We present ExtractBench, a b
Teams affected by ExtractBench 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:2607.28641v1 Announce Type: cross Abstract: We introduce the \textit{Agentic Formalism Trap} and the Evaluative Dissonance Index ($D_E$), quantifying how LLM-as-a-Judge systems conflate structural proceduralism with semantic truth under adversarial load. Analyzing 22,500 trajectories across 3 domains (GAIA, SWE-b
This lands on The Formalism Trap - 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:2607.28681v1 Announce Type: cross Abstract: Functional magnetic resonance imaging (fMRI) is a widely used technique for studying the brain. Recent methods that utilize graph neural networks (GNNs) for analysis of brain functional connectivity have shown great potential for the classification of brain disorders, s
Teams affected by MPP-GNN 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:2607.28683v1 Announce Type: cross Abstract: Large language models benefit from elements in natural language, such as metaphors and analogies in training data and inference input to achieve generalisability across different domains. However, these language elements may also lead to unwanted behaviors when metaphor
Builders evaluating Metaphor-Induced Algorithmic Steering 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:2607.28695v1 Announce Type: cross Abstract: Here is the plain text version optimized for arXiv's submission form. Custom macros (like \CV and \SI) have been converted to standard text/math so they render correctly on the webpage: Evaluating the fatigue life of structural steels conventionally requires mechanical
Builders evaluating Predicting Steel Fatigue Life from Micrographs Using Physics-Informed 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:2607.28759v1 Announce Type: cross Abstract: In X-ray CT, metallic objects cause beam hardening, photon starvation, and scattering, leading to projection inconsistency, streaks, dark bands, and structural distortions that compromise clinical diagnosis and quantitative analysis. Existing metal artifact reduction (M
What's actually new here is SCMA - 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:2607.28760v1 Announce Type: cross Abstract: As image generation models scale to ever higher resolutions, global coherence, local detail, and texture fidelity become critical axes for generation quality. However, standard flow matching treats all spatial frequencies uniformly, ignoring the natural frequency hierar
This lands on WaiT for the Signal - 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:2607.28801v1 Announce Type: cross Abstract: Benchmark datasets are central to evaluating Large Language Models (LLMs), yet they are typically conceived as monolithic tasks, obscuring substantial variation in the demands of individual samples. We introduce a dataset-centric meta-evaluation framework that audits be
This lands on Benchmarks Are Not Monolithic - gauge whether it shifts what you already ship. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2607.28849v1 Announce Type: cross Abstract: Bilevel reinforcement learning (RL) is an important framework within the literature of RL that can be used to formalize various categories of problems, such as meta-learning, hierarchical task decomposition, and reinforcement learning from human feedback (RL-HF). Most o
What's actually new here is Hypergradient-based Bilevel Reinforcement Learning with Improved Sample Complexity - see if it moves anything you maintain. For researchers and builders, this is an early signal to fold into evaluation, model choice, or agent...
arXiv:2607.28896v1 Announce Type: cross Abstract: Unified audio models capable of audio understanding, audio generation and, increasingly, audio editing are proliferating rapidly. Yet a basic question about them remains unanswered: do the two heads of a unified model agree about the same audio?
Operators using related systems should check whether TORUS 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:2607.29093v1 Announce Type: cross Abstract: Metasignal is an open-source Python package for signal detection theory (SDT) and metacognitive measurement. It implements the 17 metacognitive measures evaluated by Rahnev (2025), together with the reference variables d' (perceptual sensitivity), response criterion c (
What's actually new here is metasignal - 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:2607.29180v1 Announce Type: cross Abstract: Text-to-motion generation must produce motions that are semantically correct, temporally coherent, and physically plausible. A natural approach is to first project motion data into a structured semantic space and then train a generative model within that space.
Builders evaluating MoRAE 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:2607.29337v1 Announce Type: cross Abstract: Background and Objective: Generating realistic medical images with anatomically accurate segmentation masks helps address the shortage of annotated data in medical imaging, particularly in optical coherence tomography (OCT) of mouse eyes, where manual retinal layer deli
The thing to notice is DualDiT; 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:2607.29539v1 Announce Type: cross Abstract: Standard AI-text detection benchmarks compare human-written text against text generated directly by large language models (LLMs). While prior work has shown that rewriting and paraphrasing can degrade detector performance, it remains unclear whether performance measured
The thing to notice is ARB; decide whether it changes your next build. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2607.29561v1 Announce Type: cross Abstract: Symbolic Regression (SR) aims to discover analytical equations from observational data and plays a central role in scientific modeling. While recent Large Language Model (LLM) based approaches show promise, they face two limitations.
What's actually new here is MOT-SR - 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:2607.29614v1 Announce Type: cross Abstract: The rapid adoption of deep learning models in high-risk domains has intensified the need for trustworthy Explainable Artificial Intelligence (XAI). However, objectively evaluating explanation fidelity and aligning XAI metrics with human-centered understanding remain cri
What's actually new here is A Human-Centered Validation of the Explainability-Performance Coefficient - 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:2511.22226v2 Announce Type: replace Abstract: The standard theory of model-free reinforcement learning assumes that the environment dynamics are stationary and that agents are decoupled from their environment, such that policies are treated as being separate from the world they inhabit. This leads to theoretical
At its core this is about Embedded Universal Predictive Intelligence, 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:2603.21846v2 Announce Type: replace Abstract: Explainable AI is increasingly important to scientific discovery. However, existing methods largely ignore that explanation quality is not universal: experts differ in how they assess evidence, prioritize mechanisms, and construct explanatory narratives.
The thing to notice is Shaping Scientific Explanations to Expert Perspectives with Persona-Conditioned; 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:2604.04468v2 Announce Type: replace Abstract: Evaluating retail strategies before deployment is difficult, as outcomes are determined across multiple stages, from seller-side persuasion through buyer-seller interaction to purchase decisions. However, existing retail simulators capture only partial aspects of this
Operators using related systems should check whether What Makes a Sale? Simulating End-to-End Seller--Buyer Retail changes compatibility, cost, or access requirements. For researchers and builders, this is an early signal to fold into evaluation, model...
arXiv:2604.10475v2 Announce Type: replace Abstract: Modeling household-level decisions is central to many real-world applications, including trip planning, residential mobility and migration, disaster management, etc. Existing studies primarily rely on classical machine learning models with limited predictive capacity
The thing to notice is PEMAND; 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:2605.07161v3 Announce Type: replace Abstract: AI agents are increasingly used to diagnose and mitigate failures in production systems, known as agentic Site Reliability Engineering (SRE). Current SRE benchmarks are limited to oversimplistic SRE tasks and are unfortunately hard to extend due to bespoke designs.
Teams affected by SREGym 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:2606.13316v2 Announce Type: replace Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) is a central technique for improving long-horizon reasoning in Large Language Models (LLMs). However, existing RLVR methods often encourage unnecessarily long reasoning rollouts, which can degrade reasoning coheren
At its core this is about ReSum, worth a look if that's in your workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2509.22310v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) has achieved impressive results across domains, yet learning an optimal policy typically requires extensive interaction data, limiting practical deployment. A common remedy is to leverage priors, such as pre-collected datasets or refe
Operators using related systems should check whether Adaptive Policy Backbone via Shared Network 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:2602.10429v2 Announce Type: replace-cross Abstract: AIvilization v0 is a publicly deployed large-scale artificial society that couples a resource-constrained sandbox with a unified LLM-agent architecture, aiming to sustain long-horizon autonomy while remaining executable under a rapidly changing environment. To m
At its core this is about AIvilization v0, worth a look if that's in your workflow. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
arXiv:2605.18476v2 Announce Type: replace-cross Abstract: Coding and computation remain major bottlenecks in Markov chain Monte Carlo (MCMC) workflows, especially as modern sampling algorithms have become increasingly complex and existing probabilistic programming systems remain limited in model support, extensibility
Operators using related systems should check whether AI4BayesCode changes compatibility, cost, or access requirements. For agent builders, this marks where planning, memory, or long-horizon behavior still breaks.
Security: Dependencies: Patched high and medium severity dependency vulnerabilities via pnpm.overrides ( brace-expansion , js-yaml ) ( #6639 )
What's actually new here is mem0ai/mem0 - see if it moves anything you maintain. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
Security: Dependencies: Patched high and medium severity dependency vulnerabilities via pnpm.overrides ( axios , brace-expansion , postcss , mongoose , protobufjs ) ( #6639 )
Operators using related systems should check whether mem0ai/mem0 changes compatibility, cost, or access requirements. For builders, an open release means inspectable, self-hostable code instead of a closed hosted demo.
Improvements: Onboarding suggestions: The example commands shown by openclaw mem0 config show now suggest gpt-5-mini instead of gpt-4o ( #6704 ) Security: Dependencies: Patched high and medium severity dependency vulnerabilities via pnpm.overrides ( protobufjs , axios , postcss , mongoose ) ( #6639 )
At its core this is about mem0ai/mem0, 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.
New Features: Vector Stores: Add Qdrant server-side BM25 keywordSearch() (requires Qdrant >= 1.15.2) plus payload filter indexes, so keyword search runs without a client-side BM25 dependency ( #5851 ) Bug Fixes: Core: deleteAll() now paginates through the vector store in batches of 1000 instead of listing once, so acco
Builders evaluating mem0ai/mem0 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.
Bug Fixes: Core: delete_all() now paginates through the vector store in batches of 1000 instead of listing once, so accounts with more memories than a single page (most vector stores default to ~100) had the remainder silently left behind ( #6636 ) Vector Stores: Cap Supabase search() / list() top_k at the vecs query l
Builders evaluating mem0ai/mem0 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.
OpenAI shares new results on long-standing open problems in mathematics and theoretical computer science, including advances in geometry, cryptography, and complexity.
Teams affected by Ten advances in mathematics and theoretical computer science 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.