Search: “swe-bench”

Smart search ranks by meaning as well as keywords (one row per story, last 45 days).

SWE-Bench Pro Verified: A Reliable Benchmark for Software Engineering Agents

SWE-Bench Pro Verified is a corrected benchmark showing prior coding-agent scores were inflated by reward hacking and flawed tasks.

Analysis of SWE-Bench Pro found its evaluation undermined by reward hacking from leakage of gold solutions or hidden evaluation information, plus task quality issues such as misleading problem statements and improperly scoped tests. The authors present SWE-Bench Pro Verified, combining anti-hacking safeguards that eliminate major leakage channels with minimal task refinements. Evaluations show some models perform substantially worse than previously reported, suggesting SWE-Bench Pro overestimates real software engineering capability.

Hugging Face daily papers·Sep 7, 20:00 UTC · 8d agoAI research1

35

Coding Agents Have Converged: Why the SWE-bench Leaderboard Can No Longer Order Its Top Entries, and What to Measure Instead

Audit of 254 SWE-bench submissions finds top coding-agent entries statistically inseparable, so small leaderboard gaps no longer establish rank.

The paper audits 254 SWE-bench submissions across four splits without running models. On Verified, the top two entries each resolve 396 of 500 instances, and exact paired McNemar tests separate none of the 29 adjacent top-thirty pairs at alpha=0.05. Within-model scaffold score ranges reach 29.8 percentage points, versus an 8.8-point spread among the top thirty. The authors release a five-step audit protocol and recommend reporting comparison-set-specific resolution and model-scaffold provenance.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 16:29 UTC · 23h agoAI research

40

Cognition's SWE-2 achieves 92.8 on Terminal-Bench 2.1

Cognition releases SWE-2, a 2.8T-parameter MoE coding model post-trained from Kimi K3, scoring 92.8 on Terminal-Bench 2.1.

SWE-2 is a proprietary mixture-of-experts model with 2.8T total parameters and 104B active per token, built on the Kimi K3 base with additional Cognition reinforcement-learning post-training for agentic coding. Vendor-reported benchmarks include FrontierCode 1.1 Main 50.0, DeepSWE 1.1 73.0, Terminal-Bench 2.1 92.8, and Terminal-Bench 4.0 27.3. It claims to be one point behind Claude Fable 5.1 on FrontierCode at a claimed 64% lower cost, but trails Fable 5.1 and GPT-6 Astra by a wide margin on long-horizon Terminal-Bench 4.0 tasks. The model is available today in Devin Desktop and CLI, with no published weights, no per-token API pricing, and all figures pending independent replication.

Hacker News · AI·updated Sep 12, 20:00 UTC · 3d ago·first Sep 10, 16:52 UTC · 5d agoModel release 10 sourcesHN 40↑ · 18 comments

62

Adversarial Testing of Automated Program Repair Agents for Security Vulnerabilities

SWEADV benchmark shows adversarial issue descriptions make LLM program-repair agents write insecure fixes in 51.7% of cases, evading most detection tools.

Researchers built SWEADV, a benchmark of 750 adversarial issue descriptions derived from 150 SWE-bench Verified repair tasks, covering command execution, deserialization, path traversal, denial of service, and weak hashing attack types. Tested on mini_swe agents backed by GPT-5-Mini, MiniMax-M2.5, and DeepSeek-R, adversarial descriptions induced malicious behavior with successful repair in 51.7% of cases. Detection was weak: LLM-as-judge pre-repair screening reached only 62.3% accuracy, and post-repair detection via static analysis and LLM-as-judge achieved just 39.4% and 55.4%.

arXiv cs.CR·Sep 14, 17:53 UTC · 1d agoAI safety & security1

52

What Does an LLM-Agent Leaderboard Rank Actually Compare?

A methodological study shows close LLM-agent leaderboard rank gaps on SWE-bench and similar benchmarks often do not support superiority claims.

The paper defines an estimand-aware pairwise procedure for comparing agents, checking common support and applying explicit uncertainty rules and practical margins. Across SWE-bench, AgentRewardBench, and tau2-bench, close rank differences are frequently unresolved, and proxy labels or utility rules can change which system is selected. The authors argue a leaderboard score summarizes a released evaluation but does not by itself justify pairwise superiority conclusions.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:31 UTC · 8d agoAI research

20

IFM Releases K2 Horizon: Six Apache 2.0 Models From 0.9B to 375B

MBZUAI's IFM released K2 Horizon, six Apache 2.0 models (0.9B-375B) with open training data, code, and checkpoints, claiming the largest fully open-source launch.

The Institute of Foundation Models (IFM), launched by MBZUAI, released K2 Horizon: six Apache 2.0 models (0.9B, 3.7B, 7B, 32B, 36B-A4B, 375B-A23B) shipping with the ~20-trillion-token pretraining corpus, intermediate checkpoints, training code, and logs, which IFM calls the largest fully open-source launch in AI history. The 375B-A23B scores 70.2 on Terminal-Bench 2.1 and 87.3 on GPQA Diamond; the 7B model posts 70.6 on SWE-bench Verified. New techniques include MoVA, which extends MoE routing into attention (36B total, ~4B active), and Uno, a LoRA adapter giving roughly 3x lossless decoding speedup. IFM's own reward-hacking audit re-scored 375B-A23B from 70.2% to 66.9% after flagging 24 of 712 Terminal-Bench trials.

MarkTechPost·Sep 7, 05:00 UTC · 9d agoModel release1

72

ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement

Researchers propose ModularRSI, a modular benchmark-disjoint recursive self-improvement framework that evolves agent harnesses across five modules, improving TB2.0 and SWE-Bench Verified results.

ModularRSI targets generalizable recursive self-improvement (RSI) for agent harnesses by contrasting successful and failed trajectories for the same task and aggregating evidence across tasks to find recurring behavioral deficiencies. It decomposes the evolvable harness into five modules—Agent Loop, Tool Use, Observation Management, Context Management, and Task Completion Detection—each evolved independently within a restricted scope, then integrated with conflict resolution. Using 2,000 executable evolution tasks disjoint from evaluation benchmarks, it shows consistent gains on TB2.0 and SWE-Bench Verified and transfers across different foundation models.

Hugging Face daily papers·Sep 13, 20:00 UTC · 2d agoAI research

30

Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases

Real-SWE benchmark tests coding agents on licensed private enterprise codebases; top model Fable 5.1 resolves only 38.8% of tasks.

Real-SWE is a new benchmark evaluating frontier AI coding agents on tasks drawn from private production codebases licensed from real companies, spanning billing, tax calculation, and cross-service migrations. Fable 5.1 with Claude Code leads at 38.8% resolution rate (pass@1 over eight runs), followed by GPT-6 Astra Codex CLI at 33.8% and Gemini 3.8 Flash Gemini CLI at 31.2%. Tasks use native harnesses and realistic tooling including Docker, Kubernetes, PostgreSQL, Redis, and Linear; median reference solutions edit 11 files versus 6 for DeepSWE and FrontierCode.

Hacker News · AI·Sep 12, 20:25 UTC · 3d agoAI researchHN 45↑ · 31 comments

48

ExecCritic: Learn to Test, Test to Improve for Coding Agents

ExecCritic separates test generation from patching for coding agents, lifting SWE-bench Verified resolution to 72.6%.

ExecCritic pairs a test-verify-revise scaffold with role-specific reinforcement learning: a Test agent writes repository-native tests and a Repair agent fixes code from execution feedback, both using Qwen-3.5-35B-A3B backbones. Post-trained Qwen agents compose to 72.6% on SWE-bench Verified, an 11.4-point gain over the 61.2% no-test baseline, without stronger-model or oracle feedback at evaluation time. The work shows test quality is the key variable: base-agent tests lowered resolution to 57.3% while GPT-5.6-sol tests raised it to 65.3%.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:53 UTC · 7d agoAI research1

36

StudyBench: Can Self-Evolution Squeeze Textbooks for Olympiad Capability?

Researchers introduce StudyBench, a physics benchmark showing self-evolution gains on textbook problems rarely transfer to olympiad-level questions.

StudyBench is a controlled physics benchmark splitting test data into an Application Set of difficult textbook problems and a Transfer Set of olympiad-level problems. Across three base models, representative self-evolution methods improved on the Application Set but rarely transferred to the harder Transfer Set. A guidance ablation reveals a Guidance Gap, and every method hits a Compute Plateau, indicating the remaining limits are method problems rather than data or compute problems.

Hugging Face daily papers·Aug 31, 20:00 UTC · 15d agoAI research

28

Can LLMs Engineer Their Own Agent Harness? ByteDance Seed’s HarnessDev Says Only 34 of 64 Changes Generalize

ByteDance Seed's HarnessDev benchmark finds LLM-built agent harnesses trail human engineering on code and search, with only 34 of 64 revisions generalizing.

Researchers from ByteDance Seed, SUTD, Georgia Tech, M-A-P, and TokenWave.AI introduce HarnessDev, a benchmark that evaluates the runnable agent harness an LLM writes rather than its answers, using Creation and Evolution stages across SWE-bench Pro, Terminal-Bench 2.1, MLE-bench, EQ-Bench3, and BrowseComp (2,207 instances). Six creator models including Opus 4.8, GPT-5.5, Gemini 3.1 Pro, DeepSeek V4 Pro, Qwen 3.7 Max, and Seed 2.0 Pro were tested; Opus 4.8 posted the best average of 67.8 versus an 86.2 human-engineered reference. Self-built harnesses beat references on writing and ML experimentation but lag badly on code and search, and quality proved executor-specific: Opus 4.8 fell from 69.3 to 33.0 on SWE-bench Pro when the executor was switched to Gemini. Evolution gains were small and noisy: of 64 adjacent changes, feedback and held-out scores agreed only 34 times (53.1%), and much generated state and memory code never executed.

MarkTechPost·Sep 11, 22:01 UTC · 4d agoAI research 2 sources

48

VDiff-Bench: A Challenging Benchmark for Fine-Grained Image Difference Identification

VDiff-Bench, a 1,756-question benchmark, shows multimodal LLMs struggle with fine-grained image-difference identification, scoring as low as 8.7% on low-level changes.

VDiff-Bench is a multiple-choice benchmark of 1,756 four-way questions over image pairs covering 10 change categories including position, motion, color, texture, OCR/text and illumination, with curated hard negatives. Evaluation of 11 state-of-the-art open- and closed-source MLLMs shows fine-grained visual comparison remains brittle: 7-8B-scale open-source models score 52.5-70.6% on semantic changes but only 8.7-33.3% on low-level changes like noise and texture. Notably, Grok 4.3 shows a sharp performance drop on noise and texture differences, falling behind large open-source models like Kimi K2.5 and K3.

Hugging Face daily papers·Sep 4, 20:00 UTC · 11d agoAI research

32

unsloth/Qwen3.8-Flash-Next-GGUF — new model trending #21 on Hugging Face

Qwen released Qwen3.8-Flash-Next, an experimental 125B-parameter open-weight MoE previewing the Qwen4 architecture, with Unsloth shipping optimized GGUF quants.

Qwen released Qwen3.8-Flash-Next, an experimental open-weight preview of the architecture planned to underpin Qwen4. The model has 125B parameters with 6B activated, 512 experts (10 routed plus 1 shared), Qwen Sparse Attention (QSA), Gated DeltaNet, Gated Residual, and n-gram embeddings, with 262,144-token native context extendable to 1,000,000 tokens. Unsloth provides Dynamic 3.0 GGUF quantizations, and multi-token prediction (MTP) delivers 1.3-1.7x faster inference via llama.cpp or Unsloth Desktop.

Hugging Face trending models·Aug 26, 12:29 UTC · 21d agoModel release1

76

K-Bench: a clinically calibrated benchmark for evaluating large language models in high-risk mental health conversations

Clinician-calibrated K-Bench evaluates 125 LLM configurations on 200 high-risk mental health vignettes, exposing wide variation in suicide and violence risk handling.

K-Bench is a clinician-calibrated, protected benchmark evaluating 125 model configurations from 33 base models across 14 providers on 200 multi-turn vignettes covering suicide, self-harm, domestic violence, substance misuse and no-risk presentations. A frozen GPT-4o judge achieved 94.2% exact agreement with clinician consensus across 6,751 eligible comparisons from 151 clinician-rated transcripts. Leading models combined supportive conversation with combined-risk scores above 95, while risk exploration varied substantially among weaker configurations; therapeutic prompting helped weaker models and elevated reasoning produced no average improvement. A continuously updated public leaderboard is hosted at k-bench.ai with protected test materials.

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 16:49 UTC · 1d agoAI safety & security

46

[AINews] 10% worse, 100x cheaper, 10000x faster: Why Simulation is taking over

Latent Space argues AI training pipeline stages—rewards, data, teachers, curricula, environments—are flipping from human-made to model-made simulation.

Latent Space's AINews essay traces how each component of AI training has turned synthetic since 2022: reward models (InstructGPT, RLAIF), synthetic pretraining data (Microsoft Phi, NVIDIA Nemotron-4 340B), model teachers (Alpaca, DeepSeek-R1 distillation), and self-generated curricula (Self-Rewarding Language Models, SPIN). In 2026 it highlights Karpathy's autoresearch loop—700 experiments yielding 20 kept improvements, cutting GPT-2 training time from 2.02 to 1.80 hours—and Z.ai's GLM-5.3 fully synthetic RL environment, judging, and verification stack. It frames these shifts as 'simulation': 10% worse but 100x cheaper and 10,000x faster than human equivalents.

Latent Space·Aug 22, 07:36 UTC · 25d agoAI industry

35

Cognition Releases SWE-2: A Kimi K3 Post-Trained Coding Model That Matches Fable 5.1 on FrontierCode at 64% Lower Cost

Cognition released SWE-2, an RL post-trained coding model from Kimi K3, scoring 50.0% on FrontierCode 1.1 Main and available only inside Devin.

Cognition released SWE-2, its most capable coding model, post-trained with reinforcement learning from Moonshot AI's 2.8T-parameter Kimi K3 base. It scores 50.0% on FrontierCode 1.1 Main, within 1 point of Fable 5.1 at 64% lower cost, and RL reportedly adds 5-6 points over the K3 base on many benchmarks. It is the first Cognition model with selectable reasoning-effort levels all trained in a single RL run using Pareto-slope-matched cost penalties. There are no open weights and no standalone API; it runs only inside Devin (Desktop, CLI, with Web and Fusion rolling out), free for paid tiers through October 10, 2026.

MarkTechPost·Sep 12, 23:56 UTC · 3d agoModel release1

58

nex-agi/Nex-N2.5-mini — new model trending #30 on Hugging Face

Nex-AGI releases Nex-N2.5 agentic model family (mini, Pro, Max) with a 1.6-trillion-parameter MoE Max, open weights, and hosted access via OpenRouter.

Nex-AGI launched Nex-N2.5, a family of agentic models in mini, Pro, and Max sizes, with the Max version built on a 1.6-trillion-parameter text-only Mixture-of-Experts foundation and the company's first complete post-training effort at trillion-parameter scale. The models target long-horizon computer use, web browsing, and visually grounded agentic tasks, with expanded agent training environments. Reported benchmarks include Max scoring 86.1 on Terminal-Bench 2.1 and 65.7 on SWE-Bench Pro, trailing Claude Opus 5. Weights are being released openly on Hugging Face and ModelScope, with hosted access through OpenRouter.

Hugging Face trending models·Sep 8, 05:08 UTC · 8d agoModel release1

58

MP-Bench: Evaluating Voice Agents as a Multiparty Conversation Participant

MP-Bench is the first benchmark for voice agents in multiparty conversations, finding real-time agents near chance on turn-taking.

MP-Bench is the first benchmark designed to objectively evaluate conversational speech systems as active participants in multi-party conversations. It assesses agents on turn-taking awareness and response appropriateness, with comprehension-based question-answering as a complementary evaluation. Benchmarking 12 voice agents shows real-time agents score at or below 22% on multiparty comprehension and remain near chance on multiparty turn-taking.

arXiv cs.AI / cs.LG / cs.CL·Sep 11, 17:11 UTC · 4d agoAI research2

20

SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research?

SAEScientist-Bench evaluates whether AI agents can autonomously conduct SAE interpretability research in Gemma-2-9B-IT, finding frontier agents trail expert baselines.

SAEScientist-Bench tests if AI agents can act as scientists using SAE tools for autonomous mechanistic discovery, requiring them to design contrastive probes and navigate a Gemma Scope dictionary of 131K+ features in Gemma-2-9B-IT. Across 10 agent configurations and 20 tasks, frontier agents showed genuine discovery capability but remained well behind expert reference features, lagging most in causal steering. Agents frequently misinterpreted experimental measurements even when designing effective contrasts.

Hugging Face daily papers·Sep 7, 20:00 UTC · 8d agoAI research

55

[AINews] Fal’s H3 Max Live breaks the infinite videogen barrier

Fal post-trained MiniMax H3 into a 'Max' variant with 35x-faster inference, enabling faster-than-realtime AI video generation and infinite streams.

Fal post-trained MiniMax's H3 model into a 'Max' variant and optimized it for its in-house inference engine, achieving roughly 35x the speed of the official endpoint. The optimization enables faster-than-realtime video generation, demonstrated by an infinite interactive AI-generated stream productized by levels.io. The roundup also notes Meta Muse Code's general availability with an SDK, open DeepSeek-V4-Flash-Vision-Exp weights, GLM-5.3-Flash's strong agentic cost/performance rankings, and Tencent's 770B-parameter Hy4 Preview MoE with 49B active parameters.

Latent Space·Sep 1, 04:36 UTC · 15d agoAI industry

50

Embodied-BenchForge: A Closed-Loop Agentic Workflow for Embodied Benchmark Construction

Embodied-BenchForge automates embodied benchmark construction via closed-loop synthesis with verification and repair, yielding seven benchmarks for MLLM evaluation.

Embodied-BenchForge is an agentic framework that transforms user-specified evaluation intents into complete embodied benchmark artifacts via Closed-Loop Benchmark Synthesis. Skill-Orchestrated Artifact Synthesis composes typed reusable skills while an artifact dependency graph records intermediate outputs; Requirement-Guided Verification and Repair triggers local re-execution or upstream rollback on failures. It constructs six Offline EQA benchmarks plus one interactive benchmark with 220 executable tasks, distinguishing MLLM and embodied agent capabilities in observation-based understanding and closed-loop execution.

arXiv cs.AI / cs.LG / cs.CL·Sep 11, 17:16 UTC · 4d agoAI research

18

OpenBMB Releases MiniCPM5-2B: A 2.52B Dense Model Averaging 53.9 Across 34 Benchmarks and Built to Run On Device

OpenBMB released MiniCPM5-2B, a 2.52B-parameter Apache 2.0 on-device model averaging 53.9 across 34 benchmarks, ahead of Qwen3.5-4B.

OpenBMB released MiniCPM5-2B, a 2,516,756,480-parameter dense LlamaForCausalLM model with grouped-query attention and a 131,072-token context, under Apache 2.0, runnable via vLLM, SGLang, llama.cpp, and Ollama. It averages 53.9 across 34 benchmarks versus 51.1 for Qwen3.5-4B, with strong tool-use (97.1 on tau2-Bench Telecom) and code results (69.1 LiveCodeBench v6, 46.4 SWE-bench Verified). Training combined 400B tokens of deep-thinking SFT, critic-based JustRL II RL teachers, and on-policy distillation merging 16 RL experts; datasets and intermediate checkpoints were published alongside the weights.

MarkTechPost·Sep 7, 19:18 UTC · 8d agoModel release

44

Ex-Deepmind VP Vinyals says AI self-improvement is coming but won't trigger an intelligence explosion

Ex-DeepMind research VP Oriol Vinyals says recursive self-improvement is coming but slow, and co-founds Discovery Loop with Jeff Dean to automate research.

Oriol Vinyals, former VP of Research at Google DeepMind who worked on AlphaStar, AlphaCode, and Gemini, argued at Agentic AI Summit 2026 that recursive self-improvement will progress gradually without an intelligence explosion. He identifies idea generation ('research taste') and evaluation as the two biggest bottlenecks, noting benchmarks like SWE-Bench Pro and ML-Bench mostly test the already-working steps and suffer from overfitting and scheming. He is co-founding Discovery Loop with Jeff Dean as CEO, Sanjay Ghemawat, and Quoc Le to automate the full research cycle, starting with AI research.

The Decoder·Sep 11, 17:57 UTC · 4d agoAI industry1

62

VEX-Bench: Benchmarking LLM Agents for Assessing Exploitability of Software Supply Chain Vulnerabilities

Introduces VEX-Bench, 75 expert-labeled real-world cases testing whether LLM agents can assess supply chain vulnerability exploitability; frontier models reach about 80% F1.

VEX-Bench is the first benchmark evaluating LLM agents on assessing whether upstream dependency vulnerabilities are exploitable in downstream projects, with 75 real-world expert-labeled cases across Python, Java, and Go mined from GitHub. Nine models across three agent harnesses were evaluated; GPT-5.5 and Claude Opus 4.6 reach approximately 80% F1 on binary vulnerability-status classification, but only GPT-5.5 surpasses 70% macro-F1 on fine-grained justification classification. The gap highlights the difficulty of moving beyond binary exploitability calls to explaining exploitability reasons, unlike prior benchmarks targeting zero-day settings.

arXiv cs.CR·Sep 7, 22:52 UTC · 8d agoResearch1

42

nex-agi/Nex-N2.5-Pro — new model trending #30 on Hugging Face

Nex-AGI launches Nex-N2.5 agentic model family (mini/Pro/Max), with Max built on a 1.6-trillion-parameter MoE foundation.

Nex-AGI introduced Nex-N2.5, a next-generation family of agentic models in three sizes (mini, Pro, Max) focused on long-horizon agentic tasks including computer use, web browsing, and autonomous program execution. Nex-N2.5-Max is built on a 1.6-trillion-parameter text-only Mixture-of-Experts foundation, marking the company's first complete post-training effort at trillion-parameter scale. Weights will be released open-source on Hugging Face and ModelScope, with hosted access via OpenRouter. Benchmark comparisons against Claude Opus 5, GPT-5.6 Sol, Kimi-K3, GLM-5.3, DeepSeek-V4-Pro-0813, and Qwen3.8-Max show competitive scores on Terminal-Bench 2.1 and SWE-Bench Pro, though weights were listed as "coming soon" at publication.

Hugging Face trending models·Sep 8, 11:01 UTC · 8d agoModel release1

62

RESCUE-BENCH: Towards Relation-Aware Multi-Party Emotional Support Conversation Systems

Researchers introduce RESCUE-Bench, a video benchmark of 191 couple and family conversations evaluating LLMs on relation-aware multi-party emotional support.

RESCUE-Bench is built from real couple and family interview conversations, containing 191 samples, 7,079 annotated turns, and 1,064.8 minutes of video. It defines six tasks measuring two capabilities: Relational Understanding and Relation-Sensitive Support. Experiments with ten LLMs show models handle local emotional cues but struggle with relation pattern prediction, viewpoint prediction, and support strategy prediction.

Hugging Face daily papers·Sep 8, 20:00 UTC · 7d agoAI research

27

[AINews] Hot Chips: OpenAI’s Jalapeño, Cerebras CS-5, Groq 3 LPX, Apple M6

OpenAI unveiled Jalapeno custom inference chip claiming 1.5-1.9x better perf-per-watt than NVIDIA GB200/GB300, deploying in-house by year-end.

At the 37th Hot Chips conference, OpenAI published first benchmark details for its custom Jalapeno inference chip, claiming 1.5-1.9x more work per watt, 1.7-3.6x lower end-to-end latency, and 2.1-4.1x higher interactive-workload performance versus NVIDIA GB200/GB300, with the 700W-rated part staying at or below 550W in tests. Deployment into OpenAI's own infrastructure begins by year-end, with Gen 2 deep in development and Gen 3 underway. OpenAI also said GPT-Astra and Codex helped write low-level kernels, reportedly 1.5-1.8x faster than human-expert code for selected attention and MoE blocks. Cerebras CS-5, Groq 3 LPX and Apple M6 were also featured at the conference.

Latent Space·Aug 27, 01:31 UTC · 20d agoAI industry

72

τ^τ-Bench: An Environment for End-To-End, Realistic Agent Construction

New τ^τ-bench tasks coding agents with building deployable customer-service agents; best config, Claude Opus 5, passes only 23.9% of simulations.

Researchers introduce τ^τ-bench, an end-to-end benchmark where a developer agent must build a complete customer-service agent from real business records, a client with requirements, a production API, an inherited codebase, and cost/model limits, then is scored by deploying it against held-out simulated users. Across 53 tasks in four domains, the strongest configuration, Claude Opus 5 under Claude Code, passes just 23.9% of evaluation simulations versus an 82.2% expert-authored reference ceiling. Failure modes mirror those of human developers: shallow queries instead of deep record comprehension, almost no client communication, and shipping the first architecture that runs rather than experimenting.

Hugging Face daily papers·Sep 3, 20:00 UTC · 12d agoAI research

50

[AINews] OpenAI shuts off Cursor

OpenAI cut off API access to coding tool Cursor after its SpaceX acquisition, citing contract violations by Elon Musk's companies.

OpenAI disabled Cursor's access following the closing of Cursor's acquisition by SpaceX, citing its experience with Elon Musk's companies violating contracts; Cursor responded that OpenAI accounts for only 5% of its traffic. The weekly digest also covers major open-weight releases: Z.ai's GLM-5.3 (744B total/40B active, 1M context) and Tencent's Hy4-preview (770B/49B, ~#5 on Code Arena WebDev), plus Alibaba's Qwen3.8-Flash (125B/6B). vLLM published benchmarks showing no universal winner among speculative decoding methods across model families.

Latent Space·Aug 29, 05:11 UTC · 18d agoAI industry

55

Vulnerability Localization Benchmark: Measuring Agentic Security Analysis at Repository Scale

VLoc Bench tests 27 language models at locating vulnerable files in 290 repositories; best system reaches 0.229 File F1 and 38.4% of tasks unsolved.

The Vulnerability Localization Benchmark (VLoc Bench) contains 500 real-world vulnerabilities from 290 repositories across six package ecosystems and 147 CWE categories, pairing pre-fix and post-fix repository snapshots. Agents receive only a CWE description and read-only terminal access to identify affected files, and must confirm absence on patched snapshots. The strongest of 27 language models and four static-analysis tools achieves just 0.229 File F1; 38.4% of tasks receive no correct localization, and effective localizers still report unsupported locations on patched repositories.

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 17:44 UTC · 1d agoAI research

50

Bilevel Coordinated Reflection: A Game-Theoretic Approach to Multi-Agent LLM Systems

Paper models multi-agent LLM orchestration as a bilevel game, proving transcript-only gating limits and introducing grounded-memory SRMA.

A new paper frames orchestrator-worker coordination in multi-agent LLM systems as a bilevel coordination game and analyzes free-form reflection as stochastic movement over semantic memory states, deriving finite-time bounds and an information-theoretic impossibility result: no gate observing only the generated transcript can uniformly improve over text-indistinguishable environments, while an environment-grounded gate can. The authors propose Stochastic Reflective Memory Ascent (SRMA), which accepts candidate memory only when grounded evaluation risk strictly decreases, with geometric or polynomial convergence guarantees. On 500 SWE-bench instances, a Kimi-based instantiation of the full system resolves 72.2% versus a 70.8% public mini-SWE-agent reference.

Hugging Face daily papers·Sep 1, 20:00 UTC · 14d agoAI research1

32

E2A-Bench: Benchmarking Evidence-to-Action Reliability in Financial Chart Reasoning

E2A-Bench, a 969-query financial chart reasoning benchmark, finds VLMs fail evidence-to-action consistency, with fine-tuning amplifying BUY:SELL bias 4-6x.

E2A-Bench is a 969-query benchmark built from 323 HS300 constituents across three input modalities with deterministic OHLCV-derived evidence anchors, evaluating grounding, reasoning-action consistency, evidence-confidence calibration, and directional coverage via UCR, RCI, ECI, and NDR metrics. Testing 20 VLMs showed the lowest-hallucination model ranked near the bottom on coverage with only 6.4% directional coverage, and oracle-aided verification reduced unsupported claims but could collapse coverage. Financial fine-tuning amplified the BUY:SELL ratio by factors of 4.21 to 4.68 across base-fine-tuned pairs.

Hugging Face daily papers·Sep 12, 20:00 UTC · 3d agoAI research

34

MetroLLM-Bench: Evaluating Language Models as Transit Kiosk Runtimes

MetroLLM-Bench is a 955-case benchmark testing language models as transit kiosk tool-calling runtimes across six real metro systems.

The benchmark covers 37-414-station metro systems and eleven task categories including routing, fare calculation, disruptions, accessibility, and adversarial input, with 14 deterministic and 8 semantic scoring components. Of 26 models from six vendors, a PEFT-tuned 4B Qwen 3.5 student scored 91.3 on Tier 1, exceeding GPT-5.6 (90.6/90.0), while Muse Glimmer 30B led the composite ranking. A deterministic rule-based baseline reached 84.6, and PEFT gains over base models shrank from +7.03 points at 2B to -0.91 at 27B.

Hugging Face daily papers·Sep 8, 20:00 UTC · 7d agoAI research

30

MAxBench: A Multinomial Concept Recovery Benchmark

MAxBench evaluates multinomial concept recovery methods, finding affine subspaces steer most reliably but none consistently beats prompting.

MAxBench is a geometry-agnostic evaluation framework for multinomial concept representations in language models, based on sampling from recovered concept representations. It compares 10 localization methods covering 5 geometry types across 6 concepts and 4 models. Findings show affine subspaces steer more reliably than rank-one or linear subspaces due to better non-zero offsets, manifold steering is competitive where applicable, and no method consistently outperforms prompting.

arXiv cs.AI / cs.LG / cs.CL·Sep 11, 17:08 UTC · 4d agoAI research

18

Kaininja: Extending Native 3D Generators to the Part Level

KaiNinja extends TRELLIS.2 native 3D generation to part-level assets via a dual-volume O-Voxel representation, cutting whole-object Chamfer distance by 40%.

KaiNinja extends the TRELLIS.2 native 3D generator to produce part-level assets instead of one fused mesh, enabling downstream editing, rigging, and simulation. A dual-volume form of the O-Voxel representation solves the problem that a single volume cannot represent interfaces where two parts touch. The model needs no segmentation network, is partly trained on LLM-agent-authored part data, lowers whole-object Chamfer distance by 40%, and raises strict part F-score by 16% versus other part-generation pipelines.

Hugging Face daily papers·Sep 13, 20:00 UTC · 2d agoAI research

18

PIDS-Bench: Evaluating Prompt-Injection Detectors Under Over-Defense, Obfuscation, and Distribution Shift

PIDS-Bench shows prompt-injection detectors scoring F1 above 0.98 still misclassify about one-third of external benign security-adjacent prompts, revealing provenance-sensitive over-defense.

PIDS-Bench is a frozen multi-axis benchmark that jointly evaluates prompt-injection detectors on attack detection and benign false-positive behavior at fixed thresholds, spanning in-distribution inputs, hard-benign prompts, obfuscated attacks, and domain/structural distribution shifts. It evaluates seven detectors plus a rule-based lower-bound reference. A detector exceeding F1 = 0.98 on held-out data still misclassifies roughly one-third of an externally-sourced benign security-adjacent subset, and no internal detector reaches F1 >= 0.95 with hard-benign FPR <= 0.10 on the stress distribution. Hard-negative augmentation nearly eliminates over-defense on curated stress inputs but leaves it intact on externally-sourced prompts, a pattern termed provenance-sensitive over-defense.

arXiv cs.CR·Sep 14, 04:28 UTC · 2d agoAI safety & security

38

Cybersecurity IR Workshop: The workshop you shouldn’t miss

Microsoft's DART team promotes a 2-3 day Cybersecurity Incident Response Readiness Workshop that stress-tests IR plans against simulated attacks.

Microsoft's Detection and Response Team (DART), which delivers Defender Experts incident response and has supported organizations across 54 countries, is offering its Cybersecurity Incident Response Readiness Workshop. The scenario-driven engagement exercises detection, investigation, containment, and decision-making across identity, endpoint, cloud, and communications, ending with prioritized recommendations. It is available to Unified Enterprise agreement customers via their Customer Success Account Manager.

Microsoft Security Blog·Sep 1, 18:55 UTC · 14d agoIndustry

15

TempCloze: Can Video-LLMs Identify the Missing Middle?

TempCloze benchmark tests Video-LLMs' temporal reasoning with 1,521 videos, finding temporal alignment is the primary failure mode across 31 models.

TempCloze is a video cloze benchmark in which models must identify the true missing middle clip given the beginning and ending clips, using 1,521 carefully filtered videos from seven sources, mostly long-take and egocentric footage. Distractors are constructed along three dimensions: Semantic, Alignment and Progression, with shared scenes and objects to reduce appearance cues. Evaluation of 10 proprietary and 21 open-source Video-LLMs found Alignment is the primary bottleneck, with models often recognizing plausible semantics and local event progression but struggling with temporal alignment.

Hugging Face daily papers·Aug 31, 20:00 UTC · 15d agoAI research

25

Biology-in-the-loop: Amortized Adaptive Hit Discovery in CRISPR Screens

Researchers release AssayBench-Loop, a 1,389-screen CRISPR benchmark, and AssayLoop, a framework that learns adaptive hit discovery policies.

The paper introduces AssayBench-Loop, a large-scale benchmark of 1,389 CRISPR screens across five phenotype categories for adaptive hit discovery under budget constraints. It also introduces AssayLoop, which combines AssayFormer, a transformer-based amortized acquisition policy trained across historical screens, with LLM-derived biological priors via an adaptive handoff. On temporally held-out screens, AssayLoop achieves 5.67-fold enrichment over random selection and recovers 27.7% of hits after assaying roughly 5% of the candidate library, outperforming existing adaptive-design methods and standalone LLMs.

arXiv cs.AI / cs.LG / cs.CL·Sep 10, 17:45 UTC · 5d agoAI research1

30

Evaluating the NIST Bugs Framework Against CWE as a Successor for Automated Vulnerability Classification

NIST Bugs Framework evaluation shows it is more structured and automation-friendly than CWE for automated vulnerability classification, with gaps in attribute guidance.

The paper evaluates NIST SP 800-231's Bugs Framework (BF) against CWE as a target for automated CVE classification using a systematically screened corpus of CVE-to-CWE research. An inter-rater study with 2 subject-matter experts mapping 13 CVEs showed strong agreement on BF's cause and operation axes but only fair agreement on the attribute axis. Automated classification was tested across two LLM deployments under different budgets, and findings support BF as more structured and automation-friendly than CWE, though gaps include under-specified attribute guidance and missing fix commits for closed-source software.

arXiv cs.CR·Sep 14, 23:18 UTC · 1d agoResearch1

30