Latest

Every published item, in order · 9845 items
Nov 6
PapersBuildersSelected13:40
Survey Maps Mechanistic Interpretability Methods for AI Safety

The survey's value lies in its proposed evaluation framework and its six-way taxonomy of research orientations, which gives structure to a field that has grown mostly through scattered techniques. Researchers and safety teams can use the released notebooks to reproduce methods hands-on rather than reading only about them. The comparison between mechanistic evidence and behavioural evaluation is a useful lens for anyone deciding whether to invest in interpretability tools over standard safety testing.

arXiv cs.LG@Fazl Barez et al. (27)Source
Tomorrow 08:00, the daily lands in your inboxTelegram channel or email. Same screening as this page, no ads.
PapersBuildersSelected12:56
Strong-to-Weak OPD Matches GRPO Accuracy With 50x Fewer FLOPs

The practical lesson is that the teacher choice decides whether OPD is worth the compute: self-distillation from a model's own GRPO checkpoint is far cheaper than rerunning reinforcement learning, while generic strong teachers are not. Practitioners training domain-specialised models can consolidate separate GRPO experts via multi-teacher OPD at a fraction of the cost. Readers comparing RL and distillation budgets will find a reusable evaluation framework here.

arXiv: DeepSeek@Zeyu Leo Liu et al. (10)Source
PapersUS vs ChinaSelected12:54
CoRE Benchmark Exposes LLM Data-Cleaning Reliance on Known Entities

Benchmark scores on data cleaning may measure memorised entity knowledge rather than repair skill, so leaderboard rankings deserve scepticism when the underlying rows are public. Teams using LLMs to clean operational data should test on renamed surrogates or fresh records before trusting reported accuracy. The released controlled tables and offline replay scoring code make it straightforward to rerun the protocol on in-house data.

Event
arXiv: DeepSeek@Jingru Yang et al. (6)Source
Chinese contentarXiv 论文提出“技术混合主义”框架,分析机器生成语言的递归语义规范化

这篇论文讲了个有意思的现象:AI 写的文本被反复引用后又变成训练数据,慢慢改变人类语言习惯。想理解 AI 对社会的长期影响可以看看。

arXiv cs.AI@Charalampos M. Liapis, Epameinondas PanagopoulosSource
PapersBuildersMulti-sourceSelected11:52
ZYFA Kernels Double FP8 Attention Speed on AMD MI355X

AMD's FP8 attention path has lagged NVIDIA's, partly because the MI355X lacks hardware tracking of matrix instruction completion, making software scheduling critical. ZYFA shows that careful kernel design can close much of that gap, and its open-source release gives engineers a drop-in option for MI355X prefill workloads. Teams benchmarking AMD accelerators against H100 or B200 deployments, including those running DeepSeek-style MLA, should test it against their current AITER setup.

Event
arXiv: DeepSeek@Quentin Anthony et al. (4)Source

Only the latest 2000 items are listed; for older ones see the daily archive