代理工具

Agent、MCP、coding agent、工作流自動化與評測。

Quipu Changes Agent Writes to Knowledge Graphs from Accept-Then-Clean to Validate-Before-Commit, with Rejected Records Replayable Across Versions

Quipu places data, trust labels, rules, and decisions on dual timelines, and checks candidate state before committing agent writes. In the authors’ injected-defect test, all six defects persisted without a gate, compared with zero under Quipu, though the evidence still comes mainly from a single deterministic evaluation designed by the system’s authors.

2026-08-18

AI 安全

CompoSkill Shows Agent Skills That Pass Individual Scans Can Still Form Risk Chains at Rates of Up to 80% When Composed

CompoSkill reframes skill security from evaluating individual packages to analyzing paths involving cross-skill data flows and side effects. Across its 1,140 test cases, the highest risk-chain formation rates for white-box and black-box attacks reached 83.3% and 80.6%, respectively, but these figures are not equivalent to real-world compromise success rates.

2026-08-18

代理工具

ClawGym II Captures Trajectories at Agent Framework Boundaries, Enabling Even Black-Box Claude Code to Participate in Reinforcement Learning

Rather than modifying agent framework internals, ClawGym II intercepts calls at the model-serving boundary and reconstructs multi-turn trajectories using a prefix tree. After training, Qwen3-30A3B’s Pass@1 on ClawGym-Bench increased by 9.98 percentage points with OpenClaw and 14.81 percentage points with Claude Code.

2026-08-18

AI 安全

Compliance Detectors Still Mostly Read the Scenario Even When Given Rules; Cross-Rule Tests Fall Nearly to Chance

A new study finds that several guard models and activation probes show almost no change in judgment accuracy when rules are removed, shuffled, or replaced. In cross-rule tests designed to eliminate textual shortcuts, fast detectors perform near chance; only models using step-by-step reasoning clearly recover the ability to combine rules with scenarios.

2026-08-18

代理工具

Ten Thousand Agent Communities Exhibit Three Phases—Consensus, Polarization, and Apathy—With the Ising Model Predicting Opinion Evolution

A Stanford team simulated more than 10,000 communities of language-model agents and found that, over multiple rounds of communication, initially weak opinions gradually develop into consensus or polarization. Using an extended Ising model, the researchers predicted individual trajectories on previously unseen communication graphs. Political questions, however, also produced a rightward shift, showing that collaboration does not necessarily eliminate model bias.

2026-08-18

AI 安全

JailbreakSkill Packages Jailbreaking Strategies as Evolvable Skills, Raising HarmBench ASR by Another 13.4 Percentage Points

JailbreakSkill packages prompt rewriting, failure analysis, and skill evolution behind a unified interface, enabling red-team agents to generate new procedures from unsuccessful attack trajectories. Its second stage raises the cross-model average attack success rate on AdvBench and HarmBench by a further 17.5 and 13.4 percentage points, respectively, although evaluation relies on a GPT-4o judge.

2026-08-18

推論系統

ParaTempo Asynchronously Prunes Reasoning Branches Using Temporal Confidence, Cutting Qwen3.5 Latency by 21.8%

ParaTempo periodically probes the provisional answer distributions of individual reasoning paths, then independently prunes, retires early, or forks them. On four benchmarks, Qwen3.5-35B-A3B averaged 1.1 percentage points lower accuracy than fixed 16-path self-consistency, while reducing latency by 21.8% and generated tokens by 30.3%.

2026-08-18

代理工具

PACE-Bench Breaks Existing Agent Designs After Physics Rules Change; Best Full-Scale Configuration Reaches Just 35.9% Pass@2

PACE-Bench uses 144 executable physics environments to test whether agents can repair previously successful program designs that fail after changes to friction or material constraints. Experiments show that reflection outperforms unverified self-revision, while long-term memory can anchor stronger models to early solutions.

2026-08-17

CForce 以後期去噪結果監督早期預測，LLaDA2.1 每次前向處理 token 數由 6.94 升至 9.08

CForce 沿用擴散語言模型自己的生成軌跡，把後期較完整的預測蒸餾回早期遮罩狀態。LLaDA2.1-mini 在提高平行解碼量的同時維持準確度，但證據目前集中於同一模型家族與作者控制的評測。

2026-08-17

sMuon Projects Muon Into the LoRA-Expressible Subspace, Runs About 30% Faster Than Riemannion at the Same Loss

sMuon first projects the orthogonalized full-weight update direction into the low-rank space expressible by LoRA factors, then computes the update using only matrix multiplications. In ReLoRA experiments, it matches Riemannion’s lowest validation loss, although cross-model fine-tuning results depend heavily on the optimizer originally used for the base model.

2026-08-17

代理工具

ScienceFlow Adds Executable Workspaces to Agent Memory, Reaching a 70.22% Any-Medal Rate on 75 MLE-bench Tasks

ScienceFlow saves more than conversation summaries: it packages code, validation evidence, resource records, and workspace snapshots into recoverable stages. The official results report a 70.22% Any-Medal rate on the full MLE-bench, though cross-system comparisons remain affected by differences in models, hardware, and time budgets.

2026-08-17

Intern-S2-Mobius Shares FFN Memory Globally, With the 35B Model Reporting Nearly 4× End-to-End Inference Acceleration

Intern-S2-Mobius separates the knowledge storage and attention-based reasoning traditionally tied to each Transformer layer, allowing multiple Reasoners to repeatedly query the same global Memory. The open weights can be loaded with LMDeploy, vLLM, or Transformers, but the nearly 4× acceleration remains a result from the official evaluation under specific conditions.

2026-08-17