Song @shyoyhs
Working on machine translation at NICT. DC1/若手研究. Ski🏔️/Mountain⛰️/Marathon🏃 shyyhs.github.io Kyoto Joined March 2016-
Tweets582
-
Followers293
-
Following281
-
Likes5K
📢 Test Set RELEASED! 🚀 The test set for the #WMT25 Shared Task on QE-informed Segment-level Error Correction is now LIVE! It's time to put your MT error correction / APE methods to the test. Let's see how well they can correct machine translation! #NLProc #MT #WMT2025
We have an English--Indic Document Translation Task in Workshop on Asian Translation (WAT) 2025, which will be held together with AACL-IJCNLP 2025. Please consider participating in it 😁 sites.google.com/view/indic-doc
It's possible that digging deeper into the limited amount of existing data points equals (in theory) having more data points. youtu.be/3d6DsjIBzJ4?si…
また面白い記事を書きました。ぜひ読んでみてくださいね。 joisino.hatenablog.com/entry/anna
Thinking about explicitly exploiting searching algorithms (some kind of combinations of BFS, DFS, bidirectional search, A* for different scenarios) for increasing test time computation.
Thinking about explicitly exploiting searching algorithms (some kind of combinations of BFS, DFS, bidirectional search, A* for different scenarios) for increasing test time computation.
Reading through the technical report, really like the idea of combining an efficient model to broadly explore the search space by width, and a strong model to drill down.
Reading through the technical report, really like the idea of combining an efficient model to broadly explore the search space by width, and a strong model to drill down.
🚨Multilingual LLMs, finetuned only on English reasoning data, can still reason when asked non-English questions, showing reasoning traces that go back & forth between languages. I had so much fun working on this project Please give our paper a read! arxiv.org/abs/2505.05408
🚨Multilingual LLMs, finetuned only on English reasoning data, can still reason when asked non-English questions, showing reasoning traces that go back & forth between languages. I had so much fun working on this project Please give our paper a read! arxiv.org/abs/2505.05408 https://t.co/0wDJ7UHAjl
The PrahokBart models are finally public! Please check them out! huggingface.co/nict-astrec-at… huggingface.co/nict-astrec-at… These are the SOTA open source models for Khmer NLP! @hour_kaing @shyoyhs @MASAOUTI
The PrahokBart models are finally public! Please check them out! huggingface.co/nict-astrec-at… huggingface.co/nict-astrec-at… These are the SOTA open source models for Khmer NLP! @hour_kaing @shyoyhs @MASAOUTI
In The Cognitive Revolution, there was an interview with Jack Roe, the technical lead for reasoning and inference-time scaling on Gemini 2.5. He was asked some fairly in-depth questions during the podcast, and below are my notes and reflections based on that conversation.…
Randomly came across this classic RNN post 10 years ago karpathy.github.io/2015/05/21/rnn… "Currently it seems that word-level models work better than character-level models, but this is surely a temporary thing." Feels like there should be strong character-level LLMs
Learning optimal transport theory, which is way deeper than I expected… github.com/user-attachmen… Hoping to apply it to multilinguism someday
I'm excited to announce my new paper from my NICT internship: IteRABRe: a simple iterative pruning method that reduces model size while preserving quality with minimal computational resources. arxiv.org/abs/2503.06291 @prajdabre1 @AlhamFikri
🚨🚨🚨Excited to share our latest work: "Pralekha: An Indic Document Alignment Evaluation Benchmark", focusing on document-level alignment across 11 Indic languages. Paper: arxiv.org/abs/2411.19096 Github: github.com/AI4Bharat/Pral… Hugging Face 🤗: huggingface.co/datasets/ai4bh… 🧵1/N
We were honoured to host Dr. @ylecun at @ai4bharat! His visit was inspiring, filled with insightful discussions on our open-source efforts to advance AI for India. Moments like these highlight what makes @ai4bharat a special place to work. We’re grateful for his time with us!
I’m excited to share that our work has been accepted at COLING 2025! A huge thank to all my amazing collaborators for their contributions. @prajdabre1 @shyoyhs @Hien_UET Hideki Tanaka, @masaoutiyama
I’m excited to share that our work has been accepted at COLING 2025! A huge thank to all my amazing collaborators for their contributions. @prajdabre1 @shyoyhs @Hien_UET Hideki Tanaka, @masaoutiyama
On a lighter note: We had a paper accepted to AMTA on instruction fine tuning and synthetic data for structured document translation. Work done with @shyoyhs as a part of the @NICT_Publicity-@SAP collaboration. What was cool was the quality of reviews. Super detailed and high…

Mamoru B Komachi @mamoruk
14K Followers 7K Following Professor (Graduate School of Social Data Science) at Hitotsubashi University.
ごりちゃん🦍 @prd_xxx
5K Followers 3K Following 競プロやってるゴリラ / 競プロオンサイトによく現れる / ごりちゃん / AtCoder (Python) Algo:💧1400↑ Heu: 🐳1900↑ / IIDX @prd_iidx
のぶ @nobug5c9
386 Followers 256 Following 京大黒橋研 → NEC データサイエンスラボラトリ / 日本語 DeBERTa 学習 / KWJA & rhoknp 開発
Xincan Feng @xincanfeng
986 Followers 865 Following PhD student / applied mathematician / knowledge graph, LLMs, audio, medical, chatbot / JST SPRING / looking for position after graduation
tatyam @tatyam_prime
6K Followers 4K Following つよつよ競技プログラマーです。 音ゲーもしています。 素数大富豪をしていました。 (ノ) - ω - (ヾ)モチモチ https://t.co/UpnjW423ML https://t.co/LTnQktF6Wb icon:@MelvilleTw
けんちょん @drken1215
21K Followers 10K Following 数学やプログラミングを頑張る中学生・高校生を応援したい/開成卒 → UT数理情報卒 → 数理システム(顧問)・モノグサ/競プロ (AtCoder highest 2503, 2019年3月)/「けんちょん本」著者/Qiita 5 位/JOI 育成強化委員/ボルダリング/旅好き/ENFJHANGYO, Masatsugu @mhangyo
2K Followers 930 Following 幕張あたりの某気象会社。昔はゼロ照応の研究してました。コーパスとか機械学習好きです。 家訓は「世の中は皆さんが思っているより深くて不思議で面白い」
ミドリムシ+ @Euglenese
2K Followers 2K Following 競プロと音ゲーが好きな人間です。AtCoder(June_boy)とかウニ(虹下位)とか。浮上はプロコンの時と気が向いた時に。
シバニャン @_6v_
5K Followers 6K Following バックエンドエンジニャ よろしくにゃ 証明写真作るやつ: https://t.co/efikX7UEx7 避難先 https://t.co/jlr6MmN60N…
Katsuhito Sudoh (ja) @katsuhitosudoh
4K Followers 2K Following 奈良女子大学 教授(所属は生活環境科学系). 機械翻訳の研究をしている気がします/ Keywords: Eマウント,平SFC,平JGC,DL Gold,万年筆,インク,お茶,ZFS / ポストの内容は当人個人の見解です English: @katsuhito_sudoh
Raj Dabre @prajdabre
15K Followers 1K Following Senior Research Scientist - @google, Adjunct Faculty - @iitmadras, @iitbombay, Ex: @NICT_Publicity Use of my tweets without permission ➡️ legal action
Yuyang DONG @touuyou001
557 Followers 395 Following Chinese, Ph.D., SB Intuitions, ex-NEC (contributed to cotomi LLM models), 投資と不労取得に興味
Takafumi Morita @tm_fantasista
1K Followers 859 Following 東京大学大学院情報学環 助教 @xhi_ lab/Assistant Professor /DC2/ACT-X加速/学ロボ’17リーダー(Rits)/未踏’20スパクリ/クマ財団4期/HydroMod(CHI’22)/see-saw(Ars Electronica’22)/InflatableMod(CHI’23)
Rishu Kumar @rishdotuk
738 Followers 1K Following Long document understanding, Multilingual Evals and efficient models mainly, but other #NLProc applications in free time | vim enthusiast
Boxuan Lyu @lyu_boxuan
361 Followers 528 Following Science Tokyo 奥村・船越研究室 D1 (self-funded)/NLP/Machine Translation/Intern @ELYZA_inc → @NEC_jp_pr → @mantra_ja → @NICT_Publicity
Benjamin Marie @bnjmn_marie
1K Followers 201 Following Independent AI researcher (LLM, NLP). My blog, The Kaitchup - AI on a Budget: https://t.co/tyXQ2R8xgV
Yahui Fu @fuyahui5
24 Followers 134 Following Program-specific researcher at Speech and Audio Processing Lab.@univkyoto
Teruaki Oka @teruaki_oka3
219 Followers 190 Following つれづれなるままに。分野の殿を往く後顧学者。 SB Intuitions の評価担当。オタク。 小説書くのがうまくなりたい→いろんなテキストデータが見られる という理由で今、NLPer。 結構、何でも屋。日本語は収束せずにでたらめに使われて発散してこそ楽しい派。
Genki Kusano @genksn4
466 Followers 254 Following 仙台住み。NEC データサイエンスラボ。推薦システム、ユーザ理解、スモールデータ解析、自然言語処理が好き。学生時代はパーシステントホモロジーとカーネル法。
arlet @arlet_nephin
26 Followers 1K Following
LEO @globerleo
3K Followers 7K Following 教育行政(教育委員会)経験ありの小学校教員です。心身を壊し一度退職をして現在は再び教員をしています。それもあり、ウェルビーイング(持続的幸福)教育の研究をしています。 よろしくお願いします。 英検1級合格目指し英語の学習を続けています。あきらめと希望の闘い。
SF @S_F_tt
20 Followers 316 Following
SHITO RYO @Dlexus4
254 Followers 465 Following Hitotsubashi Komachi lab D1 | (Japanese) Semantic Change | Zh Ja En
Hayato Futami @emonosuke
367 Followers 552 Following Research engineer, Speech and language processing. Views are my own.
LM8 @ShengzheLi
72 Followers 162 Following
Yo Ehara @yo_ehara
7K Followers 3K Following Assoc. Prof., Tokyo Gakugei University. PRESTO, ACT-X, & ACT-I Researcher. EduNLP, AIED, EDM. Tweets are my personal opinions. For English accnt, cf. @EharaLab
Yashasvi Mantha @YashasviMantha
98 Followers 238 Following Mostly about tech and cinemas along with a touch of politics and a side of Space. Data Scientist | NLP - Researcher Personal opinions only.
Aditya Joshi @aadi_joshi
1K Followers 1K Following Senior Lecturer in #NLProc at @UNSWCOMPUTING . PhD from @iitbombay . 🇦🇺🇮🇳 🏳️🌈
Girish Koushik @girkosh
236 Followers 339 Following AI Researcher | PhD Student @UniOfSurrey | Multimodal Understanding | Chess Grandmaster ♟️👑
Abdul Kadir @okabdulk
66 Followers 540 Following I may not be smart but I am alive, that puts me ahead of 100 billion humans already and I guess that's enough for this life.
Pratinav Seth @pratinavseth
798 Followers 5K Following AI explorer Interpretability, Alignment, Optimization, Safety & More at AryaXAI | AI for Social Good | AAAI UC 23 Scholar | Prev. @ Mila,Bosch,Manipal.
Mark @FerjapAi
0 Followers 329 Following
Mohammed Safi Ur Rahm... @SafiKhan2k
171 Followers 878 Following PhD @ai4bharat, @iitmadras, @WSAI_IITM
h3yFr13nd @h3yFr13nd
34 Followers 498 Following
Sanjay Suryanarayanan @sanjaysury_
81 Followers 267 Following Research Associate @AI4Bharat, IIT Madras.
MKI @mki028
157 Followers 2K Following certified mediocre. NLP/CL & アニメーション enjoyer. research interests: language, learning, interpretability, or everything in-between. PhD-ing @mbzuai
lee_tutor(微信ID:... @WanjalaBri84022
84 Followers 712 Following 5年老牌工作室,300多名海外名校硕士博士写手团队! 专业代写各科 essay,thesis,assignment, coursework,groupwork, homework,reports, ☘️100%原创 ☘️修改到满意 微信:【zhue_paper】
Air Katakana @airkatakana
31K Followers 1K Following based postdoctoral researcher in ai. on trial in south korea for writing true things (see highlights). also building japanese language learning app @sottaku_app
Harsh Desai @dreamerharsh
1 Followers 7K Following
Rupaiya Bhaiya @rupaiyabhaiya
0 Followers 87 Following
MizumotoAtsushi @MizumotoAtsushi
7K Followers 8K Following Atsushi Mizumoto, Ph.D. in Foreign Language Education | Professor, Kansai University, Osaka, Japan https://t.co/udIJWp0X87 https://t.co/ySQ0IWAqFS
daniel wilson @geoframeai
1K Followers 2K Following founder and CEO of https://t.co/0aZ2rQCOYy | PhD | linguist | AI will speak every language, and we will teach it
Technoguy @Technog54494173
47 Followers 1K Following
Hiroyuki Deguchi @de9uch1_
444 Followers 362 Following Machine Translation, kNN, Decoding, Efficiency @ NTT CS lab ex. NAIST, NICT, Mantra, NII/LLMC, Ehime Univ. Gentoo / LISP / Rust / bebop jazz pianist
Mamoru B Komachi @mamoruk
14K Followers 7K Following Professor (Graduate School of Social Data Science) at Hitotsubashi University.
ごりちゃん🦍 @prd_xxx
5K Followers 3K Following 競プロやってるゴリラ / 競プロオンサイトによく現れる / ごりちゃん / AtCoder (Python) Algo:💧1400↑ Heu: 🐳1900↑ / IIDX @prd_iidx
のぶ @nobug5c9
386 Followers 256 Following 京大黒橋研 → NEC データサイエンスラボラトリ / 日本語 DeBERTa 学習 / KWJA & rhoknp 開発
Heiga Zen (全 炳河... @heiga_zen
9K Followers 204 Following Principal Scientist (Director) @GoogleDeepMind / GDM Tokyo site lead.波瀬小⇒一志中⇒鈴鹿高専⇒名工大 (IBM TJ Watson intern)⇒東芝欧州研⇒Google (Speech🇬🇧⇒Brain🇯🇵) ⇒GoogleDeepMind
Xincan Feng @xincanfeng
986 Followers 865 Following PhD student / applied mathematician / knowledge graph, LLMs, audio, medical, chatbot / JST SPRING / looking for position after graduation
tatyam @tatyam_prime
6K Followers 4K Following つよつよ競技プログラマーです。 音ゲーもしています。 素数大富豪をしていました。 (ノ) - ω - (ヾ)モチモチ https://t.co/UpnjW423ML https://t.co/LTnQktF6Wb icon:@MelvilleTw
けんちょん @drken1215
21K Followers 10K Following 数学やプログラミングを頑張る中学生・高校生を応援したい/開成卒 → UT数理情報卒 → 数理システム(顧問)・モノグサ/競プロ (AtCoder highest 2503, 2019年3月)/「けんちょん本」著者/Qiita 5 位/JOI 育成強化委員/ボルダリング/旅好き/ENFJHANGYO, Masatsugu @mhangyo
2K Followers 930 Following 幕張あたりの某気象会社。昔はゼロ照応の研究してました。コーパスとか機械学習好きです。 家訓は「世の中は皆さんが思っているより深くて不思議で面白い」
ミドリムシ+ @Euglenese
2K Followers 2K Following 競プロと音ゲーが好きな人間です。AtCoder(June_boy)とかウニ(虹下位)とか。浮上はプロコンの時と気が向いた時に。
Ryohei Sasano @cacaho
1K Followers 87 Following Researcher in Natural Language Processing / Hobbies: juggling, piano, skiing, etc. / An author of https://t.co/iSsUI21wAI
シバニャン @_6v_
5K Followers 6K Following バックエンドエンジニャ よろしくにゃ 証明写真作るやつ: https://t.co/efikX7UEx7 避難先 https://t.co/jlr6MmN60N…
Katsuhito Sudoh (ja) @katsuhitosudoh
4K Followers 2K Following 奈良女子大学 教授(所属は生活環境科学系). 機械翻訳の研究をしている気がします/ Keywords: Eマウント,平SFC,平JGC,DL Gold,万年筆,インク,お茶,ZFS / ポストの内容は当人個人の見解です English: @katsuhito_sudoh
麺匠たか松【京... @takamatu_mensho
9K Followers 12 Following ●つけ麺・らぁ麺『麺匠たか松』のアカウントです●2011年京都・四条烏丸で創業.石臼挽き全粒粉を使用した細麺と鶏魚介スープの「つけ麺」が人気のお店.季節ごとに登場する限定商品も必見です!●\★☆6/20(金)『BiVi二条店』オープンしました☆★/
Yasuaki Kakehi 筧 �... @xhi_
5K Followers 499 Following Researcher, Artist / Professor@UTokyo / 東京大学総長特任補佐 / 大学院情報学環 教授 / xlab UTokyo @xlab_utokyo / Creative Futurists Initiative @cfi_utokyo / Sustainable Well-being
Takafumi Morita @tm_fantasista
1K Followers 859 Following 東京大学大学院情報学環 助教 @xhi_ lab/Assistant Professor /DC2/ACT-X加速/学ロボ’17リーダー(Rits)/未踏’20スパクリ/クマ財団4期/HydroMod(CHI’22)/see-saw(Ars Electronica’22)/InflatableMod(CHI’23)
Sumanth @sumanthd17
4K Followers 2K Following Building Models @sarvamai PhD’ing @iitmadras @AI4Bharat, Google PhD Fellow, Past life - @GoogleAI @Mila_Quebec @IIITSC
Rishu Kumar @rishdotuk
738 Followers 1K Following Long document understanding, Multilingual Evals and efficient models mainly, but other #NLProc applications in free time | vim enthusiast
Yuyang DONG @touuyou001
557 Followers 395 Following Chinese, Ph.D., SB Intuitions, ex-NEC (contributed to cotomi LLM models), 投資と不労取得に興味
Boxuan Lyu @lyu_boxuan
361 Followers 528 Following Science Tokyo 奥村・船越研究室 D1 (self-funded)/NLP/Machine Translation/Intern @ELYZA_inc → @NEC_jp_pr → @mantra_ja → @NICT_Publicity
Benjamin Marie @bnjmn_marie
1K Followers 201 Following Independent AI researcher (LLM, NLP). My blog, The Kaitchup - AI on a Budget: https://t.co/tyXQ2R8xgV
Pratinav Seth @pratinavseth
798 Followers 5K Following AI explorer Interpretability, Alignment, Optimization, Safety & More at AryaXAI | AI for Social Good | AAAI UC 23 Scholar | Prev. @ Mila,Bosch,Manipal.
Internet of Shit @internetofshit
406K Followers 236 Following screw it, put a chip in it. say hello: [email protected]
Andrej Karpathy @karpathy
1.4M Followers 1K Following Building @EurekaLabsAI. Previously Director of AI @ Tesla, founding team @ OpenAI, CS231n/PhD @ Stanford. I like to train large deep neural nets.
Kodama @kodama26985649
186 Followers 67 Following Researcher@NII ← Kurohashi-Chu-Muwaraki lab@Kyoto Univ. NLP / Dialogue System / LLM
Kotoba Technologies @kotoba_tech
3K Followers 114 Following GenAI company breaking language barriers. 言語の壁を崩す生成AIの会社です。 📱 iOS app (🇯🇵🇺🇸🇨🇳🇰🇷): https://t.co/gKRvpsQeY6 👉 Discord: https://t.co/URPH60nLS4
NLPコロキウム @nlp_colloquium
2K Followers 72 Following 自然言語処理 (NLP)・計算言語学 (CL) の最先端の研究を著者ご本人より20分で紹介🚀 日本時間水曜正午より。毎月1〜2度開催。 Zoom URL はメーリングリストから配信します → https://t.co/qE1zbBN5SQ
Ryuichiro Higashinaka @RHigashinaka
401 Followers 1 Following
Yahui Fu @fuyahui5
24 Followers 134 Following Program-specific researcher at Speech and Audio Processing Lab.@univkyoto
Genki Kusano @genksn4
466 Followers 254 Following 仙台住み。NEC データサイエンスラボ。推薦システム、ユーザ理解、スモールデータ解析、自然言語処理が好き。学生時代はパーシステントホモロジーとカーネル法。
Teruaki Oka @teruaki_oka3
219 Followers 190 Following つれづれなるままに。分野の殿を往く後顧学者。 SB Intuitions の評価担当。オタク。 小説書くのがうまくなりたい→いろんなテキストデータが見られる という理由で今、NLPer。 結構、何でも屋。日本語は収束せずにでたらめに使われて発散してこそ楽しい派。
Miori Sagara @miorisagara
2K Followers 1K Following バオバブという会社をやってます。 https://t.co/zmqMKraAk8 東京都渋谷区生まれ横浜市中区育ち。今は札幌に来ています。 Bobab,Inc. Founder&CEO
EMNLP 2025 @emnlpmeeting
15K Followers 50 Following EMNLP 2025 - The 2025 Conference on Empirical Methods in Natural Language Processing, 2025 Hashtag: #EMNLP2025 Dates: November 5-9 Submission Deadline: May 19thKentaro Inui / 乾健... @inuikentaro
3K Followers 65 Following Professor, NLP Department, MBZUAI, UAE; Professor, Tohoku NLP Group, Tohoku University; NLU Team Leader, RIKEN AIP. NLP/CL/AI. 連絡先は変更ありません。下記Webページをご覧ください。
Yo Ehara @yo_ehara
7K Followers 3K Following Assoc. Prof., Tokyo Gakugei University. PRESTO, ACT-X, & ACT-I Researcher. EduNLP, AIED, EDM. Tweets are my personal opinions. For English accnt, cf. @EharaLab
Girish Koushik @girkosh
236 Followers 339 Following AI Researcher | PhD Student @UniOfSurrey | Multimodal Understanding | Chess Grandmaster ♟️👑
Aditya Joshi @aadi_joshi
1K Followers 1K Following Senior Lecturer in #NLProc at @UNSWCOMPUTING . PhD from @iitbombay . 🇦🇺🇮🇳 🏳️🌈
Yashasvi Mantha @YashasviMantha
98 Followers 238 Following Mostly about tech and cinemas along with a touch of politics and a side of Space. Data Scientist | NLP - Researcher Personal opinions only.
Zola Mahlaza @nqongeveg
115 Followers 439 Following Computer Scientist. Working on Natural Language Generation for Nguni languages.
COLING 2025 @coling2025
890 Followers 14 Following The 31st International Conference on Computational Linguistics (COLING 2025) . January 19-24, 2025
Sanjay Suryanarayanan @sanjaysury_
81 Followers 267 Following Research Associate @AI4Bharat, IIT Madras.
兵庫県知事 さ�... @motohikosaitoH
283K Followers 2 Following 兵庫県知事。対話と現場主義を徹底し、「躍動する兵庫」の実現に向けて県政改革をさらに前に進めていきたい。
内田彩 @aya_uchida
761K Followers 452 Following 声優・内田彩の公式Twitterです☺︎ 群馬県出身・ぐんま特使でもあります♪ プロフィール画像はぐんまちゃんのお友達・あおまくんです♡アクロスエンタテインメント所属
Keisuke Sakaguchi @KeisukeS_
1K Followers 429 Following Assoc. Prof. at Tohoku University, Sendai 🇯🇵. Natural Language Processing, Machine Learning, Psycho&Neurolinguistics. ex. @allen_ai @jhuclsp @NAIST_MAIN_EN
岡部洋一 @__obake
15K Followers 4K Following 2011年になって、Twitterなるものを触ってみる。放送大学前学長。2016年から放送大学生。obake は okabe のアナグラム。 https://t.co/0IDfOm3DC1
Haryo @haryoaw
469 Followers 448 Following PhD Student (@mbzuai). Artificial Intelligence Scientist (NLP), Gamer, Pikachu enjoyer, and shit-poster, I guess.
グランスノー奥... @okuandibu
15K Followers 524 Following こんにちは! 関西最大級のスキー場「グランスノー奥伊吹」です。 大阪、京都、名古屋から近く、アクセス抜群!多彩な全14コースで、初心者から上級者まで誰もが楽しめるゲレンデです! 今は最新のセンターハウスが新設されて、施設面も充実&快適です♪ ぜひ皆さんで遊びに来て下さい!
あらまき @ARAMAKI
2K Followers 3K Following ソーシャル・コンピューティングと医療言語処理の研究 ときどき 油画と音楽の制作 油画 🎨 https://t.co/DTPgWXXiiZ 音楽 🎶 https://t.co/SEcjgfOvGy
wadkar @wadkar
472 Followers 561 Following Linux, Kubernetes, Software Engineering, Python, Semantics, Quantum Mechanics, Geometric/Clifford Algebra; I live in the terminal
daniel wilson @geoframeai
1K Followers 2K Following founder and CEO of https://t.co/0aZ2rQCOYy | PhD | linguist | AI will speak every language, and we will teach it