#scaling laws

38 stories · 2 in 7 days · 6 in 30 days
Stories, products and papers tagged "scaling laws", newest first.
Nov 13
PapersBuildersMulti-source14:10
Chinese contentArchSpace:记录、评审和引用 LLM 架构研究的公开平台

学术圈的架构实验常常算力不够、基线也对不上,这个平台用 scaling laws 让小实验先预测大模型的表现,还直接给了 Qwen3.5 和 OLMo3 基线。做模型架构研究的同学可以上去提交提案或复现别人被验证过的想法,不用每次都从头造轮子。

Event
arXiv cs.LG@Jie Wang et al. (11)Source
Nov 11
Chinese content分子相互作用基础模型 Bonbon 展现零样本涌现生物学能力

这篇论文把大模型的涌现现象搬到了药物分子层面:只用序列训练的 Bonbon,不做任务监督就能自动学会找结合位点、判断共价与非共价。速度比共折叠快十万倍,湿实验命中率 61.5% 相当扎实。做药物设计或分子机器学习的都该看看。

arXiv cs.LG@Adi MashiachSource
Nov 8
Nov 5
Nov 4
PapersBuildersMulti-sourceSelected04:04
Stanford AI Lab's hinted scaling laws forecast GSM8K accuracy with 5600x less compute

Predicting benchmark scores without training large models changes how labs and buyers estimate model quality before committing compute budgets. The 5600x compute saving makes accuracy forecasting accessible to academic groups and startups that cannot run full-scale training runs. Teams evaluating whether to invest in a model capability can apply this method to get defensible projections at a fraction of the usual cost.

Event
Stanford AI Lab@StanfordAILabSource
Oct 26
IndustryMulti-source01:06
Chinese contentYuchen Jin 谈新 AI 实验室:多数会失败,挑战 scaling laws 者才有机会

SWE-bench 上表现很强的 LMArena 团队创始人聊了点行业实话:抄 OpenAI 作业的新实验室大多活不下去,他押注的是那些挑战 Transformer 和下一词预测范式的团队,还点破了研究员创业的通病。

Event
Yuchen Jin@Yuchenj_UWSource
Oct 16
Oct 15
PapersOfficialSelected00:49
Apple Introduces Tri-Modal Masked Diffusion Models

Pretraining diffusion models across three modalities from scratch departs from the dominant recipe of finetuning a unimodal checkpoint, potentially simplifying how text, image and audio generation are unified in one architecture. Researchers and engineers working on discrete diffusion or multimodal models can reuse its scaling-law findings and sampling defaults. Apple's publication puts it in direct conversation with autoregressive approaches from labs like OpenAI and Google.

Primary sourceApple ML ResearchSource
Oct 14
How-toBuildersSelected11:06
Stanford CS 312 Deep Learning Alchemy Course Goes Public

Course materials are fully open, so self-learners can rent GPUs on Modal and replicate assignments without Stanford enrollment. The exam-heavy design assumes students use AI agents for experiments, then test mastery without them — a template other educators may copy. It suits practitioners who want empirical intuition for LLM training behavior rather than another from-scratch coding course.

shao__meng@shao__mengSource
Oct 8
Oct 6
Oct 1
PapersBuildersSelected19:23
AI Token Value Scaling Laws Study

This research reveals a critical threshold where AI contamination in training data becomes detrimental, challenging assumptions about data quantity versus quality. For developers and researchers working with language models, this provides concrete guidance on data curation strategies and highlights the growing importance of distinguishing between human and AI-generated text in training datasets.

Tanishq Abraham (论文推介)@iScienceLuvrSource
Sep 30
Sep 29
Sep 25
Aug 21
Aug 16
Aug 12
Aug 11
Jul 28
Jul 2
Jun 13
Jun 4
Jun 3
Jun 1
May 27
May 25
May 19
May 16
May 15
May 14