Archive
新模型、開放權重、多模態、語音、影像與模型架構。
AI 基礎設施
Rather than selecting inference nodes based solely on prefix hits, MISA-T also limits how many sessions from different workloads can occupy the KV cache concurrently. The authors measured rollout throughput gains of 43.6% to 53.3% on Step3.7 and Qwen3.6-35B-A3B, but the implementation has not yet been released.
最新模型
Rather than deciding shared experts, FFN width, and the number of activated experts separately, UniF-MoE lets each token first select shared channels and then accumulates routing probabilities according to its remaining requirements. In the authors’ DeiT and BERT experiments, it improves accuracy and MoE inference cost, but it has not yet been validated on generative large language models or distributed expert parallelism.
代理工具
A Microsoft research team compared 41 languages across 2.38 million rollouts and found that agents take different tool paths when the same task is presented in another language—even with greedy decoding. The study also shows that non-English tasks rely heavily on first translating into English, although its tools are merely symbolic calls and do not capture the errors or side effects of real-world APIs.
最新模型
ReRound uses no calibration corpus. Instead, it trains a conditional diffusion model on the model’s own weights to reconsider whether weights near the midpoint of a quantization interval should be rounded up or down. It preserves the existing low-bit format at deployment with no additional inference overhead, but each model requires several hours of model-specific reconstruction upfront.
代理工具
With GPT‑5‑mini fixed as the model and a two-hour budget, researchers enabled a debugging advisor to share execution-environment constraints across search branches, allowing all 90 AIDE runs to produce valid submissions. The results suggest that agent scaffolds still contain substantial recoverable compute, though the experiments covered only nine tabular prediction competitions.
代理工具
Researchers had a local model inspect only before-and-after Python refactoring diffs and identify 13 classes of behavioral changes across 217 Rope transformations. Maintainers accepted 12 as defects, but the model’s precision was only 0.67—too low to replace testing or static analysis.
最新模型
XPeng introduces XCoT-VLA, replacing dozens of natural-language reasoning tokens with a limited set of semantic action tokens, then using flow matching to generate continuous trajectories. Open-loop tests show lower lateral error, but no closed-loop safety evaluation has been reported, and the latency figures do not cover the full in-vehicle hardware and software stack.
AI 安全
REDAgentBench executed 1,661 attack cases across sandboxed workspace, email, browser, banking, and external file services, rather than judging harm solely from an agent’s response. When the same runs were scored using service receipts and final states, the attack success rates of all six models were 7.73 to 11.72 percentage points higher than with trajectory-only scoring.
代理工具
A study of 1,867 public repositories found that agent instruction files such as CLAUDE.md continually accumulate rules while older rules are deleted less and less often. Controlled experiments showed that retaining model-readable rationales for adding rules can eliminate 99.3% of superfluous instructions, though the results have yet to be prospectively validated in large real-world codebases.
最新模型
The winning report for MeViSv2-Audio at the eighth LSVOS breaks speech transcription, visual grounding, segmentation, and target-presence detection into a multistage pipeline, then selects a trajectory based on agreement among candidate masks. The system achieved a final score of 0.769589, although the official public leaderboard still says results are to be announced, and neither the code nor complete ablation data has been released.
代理工具
Open-Ended Optimization fixes the objective, data boundaries, interaction permissions, and evaluation, but does not prescribe how the agent should gather evidence or modify skills. With GPT‑5.5 as the optimizer, it outperformed SkillOpt- and GEPA-style workflows in most settings, but medium-strength and weaker models could not maintain the same advantage.
AI 安全
A preregistered reproducibility audit found that a vulnerability PoC that executes and triggers a crash does not necessarily reproduce the specified CVE. In 20 of the 30 evaluable cases, the same signal still appeared in the patched version, showing that agentic security benchmarks need stricter oracles than strings and exit codes.