Lune

Profiles2020-2026

Graham Neubig

Carnegie Mellon University

87Papers
56Last author
128Unique co-authors
0First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers87

  1. On the Interplay of Pre-Training, Mid-Training, and RL on Reasoning Language ModelsAuthor 2 of 3ICML 2026 · 58 citations
  2. VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain KnowledgeAuthor 5 of 6ICML 2026 · 44 citations
  3. Ambig-SWE: Interactive Agents to Overcome Underspecificity in Software EngineeringAuthor 5 of 5ICLR 2026 · 35 citations
  4. Go-Browse: Training Web Agents with Structured ExplorationAuthor 2 of 2ICLR 2026 · 30 citations
  5. Agent Data Protocol: Unifying Datasets for Diverse, Effective Fine-tuning of LLM AgentsAuthor 21 of 21ICLR 2026 · 18 citations
  6. Prompt-MII: Meta-Learning Instruction Induction for LLMsAuthor 6 of 6ICLR 2026 · 9 citations
  7. Gained in Translation: Privileged Pairwise Judges Enhance Multilingual ReasoningAuthor 4 of 4ACL 2026 · 6 citations
  8. Beyond Browsing: API-Based Web AgentsAuthor 4 of 4ACL Findings 2025 · 8 citations
  9. OpenHands: An Open Platform for AI Software Developers as Generalist AgentsAuthor 24 of 26ICLR 2025 · 7 citations
  10. Stereotype or Personalization? User Identity Biases Chatbot RecommendationsAuthor 5 of 5ACL Findings 2025 · 6 citations
  11. Better Instruction-Following Through Minimum Bayes RiskAuthor 6 of 6ICLR 2025
  12. Pangea: A Fully Open Multilingual Multimodal LLM for 39 LanguagesAuthor 10 of 10ICLR 2025
  13. WebArena: A Realistic Web Environment for Building Autonomous AgentsAuthor 12 of 12ICLR 2024 · 1,197 citations
  14. Alignment for HonestyAuthor 4 of 5NeurIPS 2024 · 82 citations
  15. Synatra: Turning Indirect Knowledge into Direct Demonstrations for Digital Agents at ScaleAuthor 9 of 10NeurIPS 2024 · 45 citations
  16. Instruction-tuned Language Models are Better Knowledge LearnersAuthor 6 of 10ACL 2024 · 11 citations
  17. Synthetic Multimodal Question GenerationAuthor 7 of 8EMNLP Findings 2024 · 5 citations
  18. PAL: Program-aided Language ModelsAuthor 8 of 8ICML 2023 · 700 citations
  19. Active Retrieval Augmented GenerationAuthor 9 of 9EMNLP 2023 · 315 citations
  20. Unlimiformer: Long-Range Transformers with Unlimited Length InputAuthor 3 of 4NeurIPS 2023 · 176 citations
  21. CodeBERTScore: Evaluating Code Generation with Pretrained Models of CodeAuthor 4 of 4EMNLP 2023 · 77 citations
  22. Mega: Moving Average Equipped Gated AttentionAuthor 5 of 7ICLR 2023 · 36 citations
  23. Why do Nearest Neighbor Language Models Work?Author 3 of 3ICML 2023 · 33 citations
  24. Execution-Based Evaluation for Open-Domain Code GenerationAuthor 4 of 4EMNLP Findings 2023 · 20 citations
  25. DocPrompting: Generating Code by Retrieving the DocsAuthor 6 of 6ICLR 2023 · 18 citations
  26. DiffusER: Diffusion via Edit-based ReconstructionAuthor 3 of 3ICLR 2023 · 8 citations
  27. Computational Language Acquisition with Theory of MindAuthor 5 of 5ICLR 2023 · 5 citations
  28. Beyond Contrastive Learning: A Variational Generative Model for Multilingual RetrievalAuthor 4 of 6ACL 2023 · 3 citations
  29. Teacher Perception of Automatically Extracted Grammar Concepts for L2 Language LearningAuthor 5 of 6EMNLP Findings 2023 · 3 citations
  30. Towards a Unified View of Parameter-Efficient Transfer LearningAuthor 5 of 6ICLR 2022 · 1,182 citations

Showing the 30 most recent indexed papers.

Dusk over the sea between two cliffs drawn in fine vertical lines