Disclaimer: Congrats of course to all the accepted NeurIPS 2023 papers! This list was mostly selected from a brief skim of best papers + oral presentations that seemed somewhat relevant in the LLM setting.

• It’s possible I missed some papers in the Oral presentations category
• This doesn’t include poster papers. I’m sure there are many great poster presentations, I just didn’t have the chance to look at all of them.

Best Paper Awards

https://blog.neurips.cc/2023/12/11/announcing-the-neurips-2023-paper-awards/

• “Are Emergent Abilities of Large Language Models a Mirage?”
  • Schaeffer et al.
  • https://arxiv.org/abs/2304.15004
• “Scaling Data-Constrained Language Models”
  • Muennighoff et al.
  • https://arxiv.org/abs/2305.16264
• “Direct Preference Optimization: Your Language Model is Secretly a Reward Model”
  • Rafailov et al.
  • https://arxiv.org/abs/2305.18290
• DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models
  • Wang et al.
  • https://arxiv.org/abs/2306.11698

Oral Presentation Papers

https://nips.cc/virtual/2023/events/oral

• “OpenAssistant Conversations - Democratizing Large Language Model Alignment”
  • Kopf et al.
  • https://arxiv.org/pdf/2304.07327.pdf
• “Toolformer: Language Models Can Teach Themselves to Use Tools”
  • Schick et al.
  • https://arxiv.org/abs/2302.04761
• “Why think step by step? Reasoning emerges from the locality of experience”
  • Prystawki et al.
  • https://arxiv.org/abs/2304.03843
• “Tree of Thoughts: Deliberate Problem Solving with Large Language Models”
  • Yao et al.
  • https://arxiv.org/abs/2305.10601
• “Visual Instruction Tuning”
  • Liu et al.
  • https://arxiv.org/abs/2304.08485