Meta tags:

Headings (most frequently used words):

news, selected, publications,

Text of the page (most frequently used words):
2025 (15), the (13), liu (10), and (10), gram (10), jiacheng (9), 2023 (9), yejin (8), choi (8), hajishirzi (8), hannaneh (7), talk (7), code (7), arxiv (7), for (7), 2024 (7), infini (7), text (5), best (5), scaling (5), main (4), conference (4), commonsense (4), noah (4), smith (4), emnlp (4), demo (4), model (4), with (4), search (4), award (4), acl (3), wang (3), colm (3), ppo (3), web (3), interface (3), language (3), trillion (3), olmotrace (3), training (3), paper (3), mini (3), publications (3), inverse (3), accepted (3), llms (3), introducing (3), llm (3), data (3), received (3), prof (3), knowledge (2), oral (2), dataset (2), vera (2), plausibility (2), statements (2), value (2), more (2), decoding (2), models (2), from (2), sewon (2), min (2), documentation (2), api (2), endpoint (2), project (2), page (2), tokens (2), pang (2), video (2), twitter (2), blog (2), tracing (2), outputs (2), back (2), internet (2), scale (2), selected (2), our (2), memo (2), trap (2), year (2), where (2), are (2), mcts (2), now (2), research (2), corpora (2), science (2), university (2), worked (2), work (2), generation (2), pretraining (2), phd (2), alisa, ximing, sean, welleck, peter, west, ronan, bras, 2022, poster, generated, prompting, reasoning, wenya, dianzhuo, general, purpose, estimation, andrew, cohen, ramakanth, pasunuru, asli, celikyilmaz, don, throw, away, your, generating, preferable, guided, monte, carlo, tree, hamish, ivison, yizhong, zeqiu, valentina, pyatkin, nathan, lambert, neurips, unpacking, dpo, disentangling, practices, learning, preference, feedback, luke, zettlemoyer, spotlight, python, package, unbounded, taylor, blanton, yanai, elazar, yensung, chen, arnavi, chheda, kothary, huy, tran, byron, bischoff, eric, marsh, michael, schmitz, cassidy, trier, aaron, sarnat, jenna, james, jon, borchardt, bailey, kuehl, evie, cheng, karen, farley, sruthi, sreeram, taira, anderson, david, albright, carissa, schoenick, luca, soldaini, dirk, groeneveld, rock, yuren, wei, koh, sophie, lebrecht, ali, farhadi, jesse, dodge, system, demonstrations, track, trailer, trillions, hao, contamination, bulletin, exact, index, see, full, list, here, submission, one, third, prizes, receives, challenge, invited, mlnlp, seminar, estimating, awarded, academic, qualcomm, innovation, fellowship, tmlr, check, out, contributed, demonstrate, strongest, trends, crystal, 机器之心, wechat, featured, efficient, engine, biggest, ever, built, into, their, multi, token, real, time, available, ai2, playground, ucsd, ibm, even, compact, version, enabling, limited, budget, stanford, nlp, tpc25, parameter, consortium, bair, berkeley, honored, cohort, rising, stars, news, linkedin, github, google, scholar, name, chinese, characters, 刘嘉程, previously, computer, illinois, urbana, champaign, julia, hockenmaier, used, facebook, natural, nlg, team, current, focus, inspecting, curating, massive, also, attribution, laws, during, have, verification, automated, theorem, proving, rlhf, student, studying, washington, fortunate, advised, hanna, 2021, ceil, ordinal,

Text of the page (random words):
jiacheng liu jiacheng liu publications blog i am a now 04 2021 ceil ordinal year phd student studying ai and llms at university of washington i am fortunate to be advised by prof yejin choi and prof hanna hajishirzi my current research focus is inspecting and curating massive text corpora for llm pretraining i also work on training data attribution llm pretraining and scaling laws for llms during my phd i have worked on commonsense knowledge generation and verification automated theorem proving rlhf and text decoding previously i received b s in computer science from university of illinois at urbana champaign where i worked with prof julia hockenmaier i used to work in facebook s natural language generation nlg team my name in chinese characters is 刘嘉程 cv google scholar github twitter linkedin news 2025 11 honored to be selected for the 2025 cohort of rising stars in data science 2025 11 infini gram mini received the best paper award at emnlp 2025 2025 09 talk at bair uc berkeley 2025 07 olmotrace received the best demo award at acl 2025 2025 07 talk at tpc25 trillion parameter consortium 2025 07 talk at stanford nlp 2025 06 introducing infini gram mini an even more compact version of infini gram enabling search in internet scale text corpora with limited budget 2025 06 talk at ibm research 2025 05 talk at ucsd 2025 04 introducing olmotrace tracing llm outputs back into their multi trillion token training data in real time now available on ai2 playground 2024 07 infini gram and ppo mcts are accepted to colm 2024 2024 01 introducing infini gram an efficient text search engine and the biggest n gram lm ever built 2023 10 ppo mcts is featured by 机器之心 on wechat 2023 10 vera and crystal are accepted to emnlp 2023 main conference 2023 09 the inverse scaling paper is accepted to tmlr check out our contributed dataset memo trap where llms demonstrate the strongest inverse scaling trends 2023 07 i am awarded the qualcomm innovation fellowship for academic year 2023 2024 2023 05 invited talk the the mlnlp seminar estimating the plausibility of commonsense statements 2023 02 our submission to the inverse scaling challenge memo trap receives one of the 11 third prizes selected publications see my full list of publications here infini gram mini exact n gram search at the internet scale with fm index hao xu jiacheng liu yejin choi noah a smith hannaneh hajishirzi emnlp 2025 main conference best paper award arxiv project page web interface api endpoint code documentation contamination bulletin olmotrace tracing language model outputs back to trillions of training tokens jiacheng liu taylor blanton yanai elazar sewon min yensung chen arnavi chheda kothary huy tran byron bischoff eric marsh michael schmitz cassidy trier aaron sarnat jenna james jon borchardt bailey kuehl evie cheng karen farley sruthi sreeram taira anderson david albright carissa schoenick luca soldaini dirk groeneveld rock yuren pang pang wei koh noah a smith sophie lebrecht yejin choi hannaneh hajishirzi ali farhadi jesse dodge acl 2025 system demonstrations track best demo award arxiv blog web interface code twitter trailer video demo video infini gram scaling unbounded n gram language models to a trillion tokens jiacheng liu sewon min luke zettlemoyer yejin choi hannaneh hajishirzi colm 2024 oral spotlight 2 arxiv project page web interface api endpoint python package code documentation unpacking dpo and ppo disentangling best practices for learning from preference feedback hamish ivison yizhong wang jiacheng liu zeqiu wu valentina pyatkin nathan lambert noah a smith yejin choi hannaneh hajishirzi neurips 2024 arxiv code models don t throw away your value model generating more preferable text with value guided monte carlo tree search decoding jiacheng liu andrew cohen ramakanth pasunuru yejin choi hannaneh hajishirzi asli celikyilmaz colm 2024 arxiv code vera a general purpose plausibility estimation model for commonsense statements jiacheng liu wenya wang dianzhuo wang noah a smith yejin choi hannaneh hajishirzi emnlp 2023 main conference oral arxiv code model demo dataset generated knowledge prompting for commonsense reasoning jiacheng liu alisa liu ximing lu sean welleck peter west ronan le bras yejin choi hannaneh hajishirzi acl 2022 main conference arxiv code talk poster