An Yan Profile
An Yan

@AnYan_ai

Followers
76
Following
179
Statuses
189

@SFResearch Prev-@UCSanDiego @Mircosoft Vision-Language.

Joined February 2023
Don't wanna be here? Send us removal request.
@AnYan_ai
An Yan
@AnYan_ai
4 months
I am attending #COLM2024 in Philly! Will present our paper “List Items One by One: A New Data Source and Learning Paradigm for Multimodal LLMs” on Monday morning ⏰ Come and chat if you are interested in multimodal LLMs, synthetic data and training recipes!
Tweet media one
1
4
24
@AnYan_ai
An Yan
@AnYan_ai
5 hours
RT @iScienceLuvr: Meta researchers used AI to predict the text a person was typing just from non-invasive brain recording! With EEG, their…
0
184
0
@AnYan_ai
An Yan
@AnYan_ai
5 days
RT @suchenzang: last walk down memory lane of 2022: but of course the godfather of AI only remembers the single f…
0
9
0
@AnYan_ai
An Yan
@AnYan_ai
5 days
RT @ArmenAgha: Only one of the teams between Zetta/LLaMa had an open-source pre-training codebase, shared datasets and experiments internal…
0
21
0
@AnYan_ai
An Yan
@AnYan_ai
7 days
RT @jxbz: I ran the "watermark" experiment which I learnt about from @norabelrose. 1000 steps of dualized training erases an "a" that was s…
0
1
0
@AnYan_ai
An Yan
@AnYan_ai
7 days
RT @littmath: Some brief impressions from playing a bit with o3-mini-high (the new reasoning model released by OpenAI today) for mathematic…
0
114
0
@AnYan_ai
An Yan
@AnYan_ai
7 days
RT @SeunghyunSEO7: The concept of critical batch size is quite simple. Let’s assume we have a training dataset with 1M tokens. If we use a…
0
90
0
@AnYan_ai
An Yan
@AnYan_ai
7 days
RT @giffmana: Now's the time!! > Install Cursor > open some horribly large and scary foreign code base, maybe HF or PyTorch or big_vision…
0
55
0
@AnYan_ai
An Yan
@AnYan_ai
14 days
RT @JJitsev: (Yet) another tale of Rise and Fall: DeepSeek R1 is claimed to match o1/o1-preview on olympiad level math & coding problems.…
0
235
0
@AnYan_ai
An Yan
@AnYan_ai
16 days
RT @natolambert: Meta is definitely not alone in this. And its normally overblown too.
Tweet media one
0
938
0
@AnYan_ai
An Yan
@AnYan_ai
19 days
RT @ClementDelangue: Current best open source video generation model?
0
8
0
@AnYan_ai
An Yan
@AnYan_ai
1 month
RT @SimonShaoleiDu: Introducing StoryEval: our new video generation benchmark! Can a model present short stories like 'How to put an elepha…
0
6
0
@AnYan_ai
An Yan
@AnYan_ai
1 month
RT @SFResearch: 🔬🔬🔬Introducing ProVision: A new system for transforming images into verified instruction data for multimodal language model…
0
34
0
@AnYan_ai
An Yan
@AnYan_ai
1 month
@m2saxon @WenhuChen dude🤣
0
0
0
@AnYan_ai
An Yan
@AnYan_ai
1 month
RT @corefpark: New paper! “In-Context Learning of Representations” What happens to an LLM’s internal representations in the large context…
0
180
0
@AnYan_ai
An Yan
@AnYan_ai
1 month
RT @XihuiLiu: Thank @_akhaliq for sharing our work! Up to 16x less inference steps and 9.5x actual speedup for autoregressive visual genera…
0
25
0
@AnYan_ai
An Yan
@AnYan_ai
1 month
RT @DimitrisPapail: I've been thinking about in-context learning for nearly 3 years. While there is still plenty I don't fully understand,…
0
90
0
@AnYan_ai
An Yan
@AnYan_ai
2 months
RT @nrehiew_: 9th highest scored ICLR 2025 paper 8,8,8,10. Worth noting all reviewers increased their scores by 2 after rebuttals tldr: t…
0
102
0
@AnYan_ai
An Yan
@AnYan_ai
2 months
RT @MinhyukSung: #DiffusionModels 🎓 Our "Diffusion Models and Their Applications" course is now fully available! It includes all the lectur…
0
98
0
@AnYan_ai
An Yan
@AnYan_ai
2 months
RT @cloneofsimo: Wait o1 mightve been this work all along?
Tweet media one
0
84
0
@AnYan_ai
An Yan
@AnYan_ai
3 months
RT @rohanpaul_ai: Academic researchers can now pre-train billion-parameter models using just 4 GPUs through optimized configurations Strat…
0
116
0