SLDBench v2发布:Claude Opus 5.5仅用4%算力成本超越经典拟合基线
jiqizhixin · x · 2026-11-16
Researchers from CUHK-Shenzhen, HKU, Tsinghua, Peking University and collaborators introduce SLDBench v2 (arXiv preprint) to benchmark active scaling-law discovery: agents see a catalog of experiments with prices, choose which results to buy, refine candidate laws, and decide when to stop. The environment replays recorded training runs at original compute cost and validates predictions on hidden larger-scale experiments. Claude Opus 5.5 reaches median R² 0.570 using just 4.01% of the pool's historical compute, beating the classic reference law fitted on all data (0.455). Across 30 tasks, all eight evaluated systems degrade when forced to actively select experiments versus receiving all outcomes — cheap observations can leave key assumptions untested.
「模型」频道最新
- Physical Intelligence 系团队发布 FLUX-mimic-x:7B 视频动作模型登顶 RoboLab 榜 — robrombach · 2026-11-16
- Qwen4-max发布页被吐槽:只有标题和评论区,无模型卡无评测 — petrusenko_max · 2026-11-16
- Grok Bot 1.0 发布:机器人可入会、主动行动并自我改进 — mattyp · 2026-11-16
- 用户实测惊呼:Qwen4 Max Preview 表现强到「不该出现在人类世界」 — rickasaurus · 2026-11-16
- 曝新模型或为 Claude 新档位:从 $10/$50 降至 $8/$40 — lyraxana · 2026-11-16
- NVIDIA Parakeet每30秒硬补句号:80%转述被无声截断 — shauntrennery · 2026-11-16