#icml — Public Fediverse posts

Linguistic Calibration trains Llama 2 to emit confidence phrases that let a downstream reader make calibrated forecasts on related questions. The key move is defining calibration through reader utility instead of self-reported probability. Hedged text that doesn't help the reader makes no forecasting progress, so generic hedging can't game the objective.

https://benjaminhan.net/posts/20260505-linguistic-calibration/?utm_source=mastodon&utm_medium=social

#LLMs #Calibration #Hallucination #ICML #AI

Conformal Factuality casts LM correctness as uncertainty quantification. Decompose the answer into sub-claims, score each, drop the low-confidence ones until the retained set is ~1-α factual. The sub-claim decomposition is doing most of the work, and the conformal machinery rides on top. Atomic-claim splitters have known failure modes, and the guarantee inherits them.

https://benjaminhan.net/posts/20260505-conformal-factuality/?utm_source=mastodon&utm_medium=social

#ConformalPrediction #Calibration #Hallucination #LLMs #ICML #AI

#CFP for `Differentiable Almost Everything: Differentiable Relaxations, Algorithms, Operators, and Simulators`, a workshop at #ICML

https://differentiable.xyz/

https://twitter.com/FHKPetersen/status/1657283011459821569

Differentiable programming is a powerful tool, so I am quite interested in this workshop (especially as a #JuliaLang user, which has fantastic #AD support).

#AutomaticDifferentiation #ML

Looking to follow (and be followed!) #mlbio #bioml #biotech #bayes #bayesian #generativemodels #startup #startups #medicine #biology #representationlearning #techbio #immunology #neurips #icml #iclr #cambridge #nyc