Do LLMs need rationales for learning from mistakes? π€
When LLMs learn from previous incorrect answers, they typically observe corrective feedback in the form of rationales explaining each mistake. In our new preprint, we find these rationales do not help, in fact they hurt performance!
π§΅
13.02.2025 15:38 β π 21 π 9 π¬ 1 π 3
New preprint out! Thrilled to share our new work led by @lisaalaz.bsky.social
13.02.2025 18:05 β π 1 π 0 π¬ 0 π 0
Check out @lisaalaz.bsky.social's internship work with us @cohere.com questioning the rationale behind rationales π₯
13.02.2025 16:18 β π 4 π 1 π¬ 0 π 0
PhD student @ ETH ZΓΌrich | all aspects of NLP but mostly evaluation and MT | go vegan | https://vilda.net
Computational Linguistics MS @ UW | Staff Data Scientist (NLP) @ Cision
ibrahimsharaf.github.io
Developer of numpad.io and meridianapp.co
stealth // Gemini RL+inference @ Google DeepMind // Conversational AI @ Meta // RL Agents @ EA // ML+Information Theory @ MIT+Harvard+Duke // Georgia Tech PhD // Ψ²Ω Ψ²ΩΨ―Ϊ―Ϋ Ψ’Ψ²Ψ§Ψ―Ϋ
π{NYC, SFO, YYZ}
π https://beirami.github.io/
Waiting on a robot body. All opinions are universal and held by both employers and family.
Literally a professor. Recruiting students to start my lab.
ML/NLP/they/she.
Assistant Professor in Computer Science at USC | NLP, ML
Visiting Scientist at Schmidt Sciences. Visiting Researcher at Stanford NLP Group
Interested in AI safety and interpretability
Previously: Anthropic, AI2, Google, Meta, UNC Chapel Hill
Head of Responsible AI, CTO Office, Bloomberg.
professor for natural language processing, head of
BamNLP @bamnlp.de
π Duisburg, Stuttgart, Bamberg
#NLProc #emotion #sentiment #factchecking #argumentmining #informationextraction #bionlp
We build secure, scalable, and private enterprise-grade AI technology to solve real-world business problems. Join us: http://cohere.com/careers
Policy at Cohere.com. Formerly @adalovelaceinst.bsky.social.
Cat is called Lola.
I lead Cohere For AI. Formerly Research
Google Brain. ML Efficiency, LLMs,
@trustworthy_ml.
AI @ OpenAI, Tesla, Stanford
Researcher in ML/NLP at the University of Edinburgh (faculty at Informatics and EdinburghNLP), Co-Founder/CTO at www.miniml.ai, ELLIS (@ELLIS.eu) Scholar, Generative AI Lab (GAIL, https://gail.ed.ac.uk/) Fellow -- www.neuralnoise.com, he/they
FR/US/GB AI/ML Person, Director of Research at Google DeepMind, Honorary Professor at UCL DARK, ELLIS Fellow. Ex Oxford CS, Meta AI, Cohere.
PhD @ Kingβs College London β’ prev CambridgeNLP, TU Wien, intern GoogleDeepmind β’ NLP, Data-centric ML, Multimodality
http://mubasharaakhtar.com
seeks to understand language.
Head of Cohere Labs
@Cohere_Labs @Cohere
PhD from @UvA_Amsterdam
https://marziehf.github.io/
Getting paid to complain about LLM Evaluation at Cohere. #NLP #NLProc
https://dennis-aumiller.de
MLE, aspiring research fairy, music & coffee lover
