Postdoc at Meta and university of Washington in NLP. Before: PhD from Charles University (Prague 🏰). Interested in going into the inner workings of neural networks 🔍, multilinguality 🌍, tokenization 🔡 and fairer NLP ⚖️ (he/him)https://bsky.app/profile/tomlim.bsky.socialhttps://bsky.app/profile/tomlim.bsky.social/post/3mlt7xbj5r22rSee you there! 🌉🔠 [contains quote post or other embedded content]14 May 2026 16:20 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3mlt7xbj5r22rhttps://bsky.app/profile/tomlim.bsky.social/post/3mkzwjnfvys26We present Compute Optimal Tokenization! 🔠 Common in LLM scaling works stick to one tokenizer, sweeping data/model size. But what happens when we control the tokenizer’s compression rate (bytes/token)? Here we sweep tokenizers, params, and data across compute budgets: [1/N]04 May 2026 14:55 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3mkzwjnfvys26https://bsky.app/profile/tomlim.bsky.social/post/3lubayfns4c2yCheck the BLT poster at @aclmeeting.bsky.social . It’s just fortaste before the main presentation at @tokshop.bsky.social next week from Artidoro Pagnoni!18 Jul 2025 20:11 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3lubayfns4c2yhttps://bsky.app/profile/tomlim.bsky.social/post/3lu7fe6sskc2sIt’d be great to meet at Tokenization Workshop @tokshop.bsky.social #icml tomorrow July 18 starting at 8:45 in Meeting 112-113! [contains quote post or other embedded content]18 Jul 2025 02:24 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3lu7fe6sskc2shttps://bsky.app/profile/tomlim.bsky.social/post/3lu2amyu24s2yI'm pleased to be in Vancouver for @ICML this week 🇨🇦🤖. I'll be happy to chat about multilingual, multimodal LMs and tokenization(free).16 Jul 2025 01:16 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3lu2amyu24s2yhttps://bsky.app/profile/tomlim.bsky.social/post/3lqnpjk3ypc2eIf you have experience with tokenization (who doesn’t) your help with reviewing will be hugely appreciated! 🔠🔡 [contains quote post or other embedded content]02 Jun 2025 21:23 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3lqnpjk3ypc2ehttps://bsky.app/profile/tomlim.bsky.social/post/3lmukzmog5c2fIt’s finally official: the long-awaited Tokenization Workshop is here! [contains quote post or other embedded content]15 Apr 2025 17:10 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3lmukzmog5c2fhttps://bsky.app/profile/tomlim.bsky.social/post/3llwewx7tc22kSo, apparently, confusing these two buttons can ignite a serious flame-war in reviewer-author discussion🔥 @aclmeeting.bsky.social03 Apr 2025 17:01 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3llwewx7tc22khttps://bsky.app/profile/tomlim.bsky.social/post/3llokom4ho22xExcited to continue my research adventure as a postdoc at @uwnlp.bsky.social and Meta! I’ve joined @lukezettlemoyer.bsky.social’s fantastic lab. Together, we plan to rethink how LLMs perceive data to unlock their capabilities to uncharted language and, further, beyond text!31 Mar 2025 14:23 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3llokom4ho22xhttps://bsky.app/profile/tomlim.bsky.social/post/3laqgrg2igs2lTokenization is so back! at #EMNLP [contains quote post or other embedded content]12 Nov 2024 08:41 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3laqgrg2igs2lhttps://bsky.app/profile/tomlim.bsky.social/post/3laopkmlszc2g#firstpost Are you working on NLP for low-resource or non-Latin script languages? If yes, I have great news for you! Our MYTE tokenizer and MyT5 models 🪲 are now easily available through🤗. It’s easy to try:11 Nov 2024 16:13 +0000at://did:plc:uulvy5pdfmvxbrxtzeibwq2x/app.bsky.feed.post/3laopkmlszc2g