🆕 Findings of #EMNLP2026 🆕
Can you tell a good tokenizer from a bad one without training a model? We tested 29 intrinsic tokenizer metrics against 159 language models in nine languages. None of them reliably predicts downstream performance across suite and language combos 🧵
We’re bringing natural-language control to any scientific generator.
We are very excited to introduce LLM-to-Flow Distillation (LFD), a method to adapt scientific generators to high-level natural-language discovery goals.
Starting from textual goals like “design peptides with promising anticancer activity” or “design small molecules binding to the D2 dopamine receptor”, LFD distills high-level goal-related knowledge from LLMs into scientific generators through iterative DPO on LLM-generated sub-goals.
🔥 LFD does not need task-specific proxy rewards, works with any frontier LLM via API-access, and outperforms state-of-the-art reward-guided methods such as FlowGRPO and DiffusionNFT on held-out computational evaluations. (1/6)
How should computational (psycho)linguists properly apply token-level language models to the field’s inherently character-level problems?
We try to bring some clarity in this new EMNLP paper 📣
Spoiler: you can (and should!) convert an LM over tokens into an LM over characters.
If you're at #ACL2023NLP, stop by the poster session tomorrow @ 16h for our paper "On the Efficacy of Sampling Adapters"! Hope to see some of you there :)
https://t.co/vtrpKyCxcV
@tpimentelms@Malagutti_Luca@weGotlieb @ryandcotterell
Come to our ACL tutorial tomorrow at 14h on generating text from language models!
Material will be online here: https://t.co/nWklOlwuIS
w/ @tpimentelms@afra_amini@johnhewtt@Malagutti_Luca @ryandcotterell