Clara Na

133 posts

Clara Na banner
Clara Na

Clara Na

@claranahhh

PhD student at @LTIatCMU / @SCSatCMU she/her, prev. @UVA and intern @ai2_allennlp @/clara on https://t.co/GHxXbrRHSB and @/clarana on https://t.co/47UIhMGaRd

Pittsburgh, PA Katılım Eylül 2021
606 Takip Edilen1K Takipçiler
Sabitlenmiş Tweet
Clara Na
Clara Na@claranahhh·
Building/customizing your own LLM? You'll want to curate training data for it, but how do you know what makes the data good? You can try out recipes👩‍🍳 iterate on vibes✨ but we can't actually test all possible combos of tweaks,,, right?? 🙅‍♂️WRONG! arxiv.org/abs/2410.15661 (1/n) 🧵
Clara Na tweet media
English
3
35
172
26.4K
Clara Na retweetledi
Sri Kundurthy
Sri Kundurthy@srkvatsa·
excited to present our paper on Spreadsheet Arena, co-led with @claranahhh and the @MeridianAgent team, at ICML next week!
Spreadsheet Arena@sheetarena

Spreadsheets have entered the arena! ⚔️ Announcing Spreadsheet Arena, the first research platform for human preference rankings on LLM-generated spreadsheets. The results? @AnthropicAI Claude Opus is on top, but the gap is tighter than you’d think. w/ @LTIatCMU, @Cornell, and @scale_ai. 🧵

English
0
1
9
734
Clara Na retweetledi
Vijay V.
Vijay V.@vijaytarian·
Continuous reward models can measure partial progress on a task, unlike binary rewards (e.g. RLVR). But we find they inevitably assign wildly different scores to equally-good responses, which can lead to bad policies. Surprisingly, dense rewards are often better if discretized!🧵
Vijay V. tweet media
English
2
19
63
13.1K
Clara Na retweetledi
Lindia Tjuatja
Lindia Tjuatja@lltjuatja·
In case you’ve been wondering what I’ve been up to these days… So excited to (re)join an amazing community of linguists and NLP researchers at UT :)
UT Linguistics Dept@UT_Linguistics

We are excited to announce that Lindia Tjuatja (@lltjuatja) will be joining us as an Assistant Professor, starting in Fall 2027! Lindia is an alum of UT Linguistics and Electrical and Computer Engineering, and is currently finishing her PhD at CMU. Welcome back to UT, Lindia!

English
44
22
295
43.7K
Clara Na retweetledi
Spreadsheet Arena
Spreadsheet Arena@sheetarena·
Spreadsheets have entered the arena! ⚔️ Announcing Spreadsheet Arena, the first research platform for human preference rankings on LLM-generated spreadsheets. The results? @AnthropicAI Claude Opus is on top, but the gap is tighter than you’d think. w/ @LTIatCMU, @Cornell, and @scale_ai. 🧵
GIF
English
2
5
37
21.1K
Clara Na retweetledi
Amanda Bertsch
Amanda Bertsch@abertsch72·
Can LLMs accurately aggregate information over long, information-dense texts? Not yet… We introduce Oolong, a dataset of simple-to-verify information aggregation questions over long inputs. No model achieves >50% accuracy at 128K on Oolong!
Amanda Bertsch tweet media
English
13
68
358
81.8K
Clara Na retweetledi
Parv Kapoor ✈️ ICML 26 /\ RSS 26
Parv Kapoor ✈️ ICML 26 /\ RSS 26@Parvkpr·
Last year, I came across the idea of constrained decoding (I know, late to the party) and was fascinated. The ability to enforce constraints for LLMs at inference time without fine tuning is a powerful idea. It got me thinking, can we do this for robot foundation models? 1/n🧵
English
1
7
33
5.3K
Clara Na retweetledi
Clara Isabel Meister
Clara Isabel Meister@clara__meister·
I've recently been fascinated by tokenization, a research area in NLP where I still think there's lots of headway! In an effort to encourage research, I made a small tokenizer eval suite (intrinsic metrics) with some features I found missing elsewhere: github.com/cimeister/toke…
English
4
15
165
10.3K
Clara Na retweetledi
Vashisth
Vashisth@vash_tiwari·
Wrapped up an incredible summer @GoogleAI, working on post training (distillation + rl) for Gemma models. Next I’ll be joining @LTIatCMU as a PhD student. Incredibly grateful to my rec writers @BeidiChen, Emma Strubell, and @gneubig for their support :))
Vashisth tweet media
English
4
3
130
11K
Clara Na retweetledi
Lindia Tjuatja
Lindia Tjuatja@lltjuatja·
When it comes to text prediction, where does one LM outperform another? If you've ever worked on LM evals, you know this question is a lot more complex than it seems. In our new #acl2025 paper, we developed a method to find fine-grained differences between LMs: 🧵1/9
Lindia Tjuatja tweet media
English
1
30
157
34.4K
Clara Na
Clara Na@claranahhh·
@lltjuatja there are no mistakes just happypillaccidents
English
0
0
4
177
Lindia Tjuatja
Lindia Tjuatja@lltjuatja·
the face of someone who has made a mistake
Lindia Tjuatja tweet media
English
2
0
31
916
Clara Na retweetledi
Mayee Chen ✈️ ICML 🇰🇷
Mayee Chen ✈️ ICML 🇰🇷@MayeeChen·
!!! I'm at #ICLR2025 to present 🧄Aioli🧄 a unified framework for data mixing on Thursday afternoon! 🔗 arxiv.org/abs/2411.05735 Message me to chat about pre/post training data (mixing, curriculum, understanding); test-time compute/verification; or to try new food 🇸🇬
Mayee Chen ✈️ ICML 🇰🇷 tweet media
English
2
47
155
19.3K
Clara Na retweetledi
Amanda Bertsch
Amanda Bertsch@abertsch72·
coming to a NAACL 2025 near you! 🌞 Looking forward to discussing with folks in Albuquerque :) The camera-ready is on arxiv now, with more models, more tasks, and more compared settings-- including results comparing ICL to full finetuning! arxiv.org/abs/2405.00200
Amanda Bertsch@abertsch72

In-context learning provides an LLM with a few examples to improve accuracy. But with long-context LLMs, we can now use *thousands* of examples in-context. We find that this long-context ICL paradigm is surprisingly effective– and differs in behavior from short-context ICL! 🧵

English
1
12
64
7.8K
Clara Na retweetledi
Jimin Mun
Jimin Mun@jiminmun_·
Check out our work on improving LLM's ability to seek information through asking better questions! 💫
Stella Li ✈️ ICML🇰🇷@StellaLisy

Asking the right questions can make or break decisions in high-stake fields like medicine, law, and beyond✴️ Our new framework ALFA—ALignment with Fine-grained Attributes—teaches LLMs to PROACTIVELY seek information through better questions🏥❓ (co-led with @jiminmun_) 👉🏻🧵

English
0
7
33
3.4K
Clara Na retweetledi
Akhila Yerukola
Akhila Yerukola@akhila_yerukola·
Did you know? Gestures to express universal concepts—like wishing for luck—vary WIDELY across cultures? 🤞means luck in US but deeply offensive in Vietnam 🚨 📣We introduce MC-SIGNS, a test bed to evaluate how LLMs/VLMs/T2I handle such nonverbal cues 📜: arxiv.org/abs/2502.17710
Akhila Yerukola tweet media
English
2
15
52
9.8K
Clara Na retweetledi
Michael Saxon
Michael Saxon@m2saxon·
🚨😱Obligatory job market announcement post‼️🤯 I'm searching for faculty positions/postdocs in multimodal/multilingual NLP and generative AI! I'll be at #NeurIPS2024 presenting our work on meta-evaluation for text-to-image faithfulness! Let's chat! Website in bio, papers in🧵
Michael Saxon tweet media
English
6
60
214
38.6K
Kayo Yin
Kayo Yin@kayo_yin·
it’s interesting how different my voice sounds in different languages
English
5
2
95
12.7K
Clara Na retweetledi
Shuyan Zhou
Shuyan Zhou@shuyanzh36·
My lab at Duke has multiple Ph.D. openings! Our mission is to augment human decision-making by advancing the reasoning, comprehension, and autonomy of modern AI systems. I am attending #emnlp2024, happy to chat about PhD applications, LLM agents, evaluation etc etc!
English
8
55
249
35.4K
Clara Na retweetledi
Lindia Tjuatja
Lindia Tjuatja@lltjuatja·
💬 Have you or a loved one compared LM probabilities to human linguistic acceptability judgments? You may be overcompensating for the effect of frequency and length! 🌟 In our new paper, we rethink how we should be controlling for these factors 🧵:
Lindia Tjuatja tweet media
English
3
18
138
29.7K

Keşfet

@MeridianAgent @AnthropicAI @LTIatCMU @Cornell @scale_ai @GoogleAI @BeidiChen @gneubig