Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Diffusion Models are Efficient Data Generators for Human Mesh RecoverySenseTime · IEEE Transactions on Pattern Analysis and Machine Intelligence16 citationsJan 1, 2025Prototypical Human-AI Collaboration Behaviors from LLM-Assisted Writing in the WildMicrosoft · Conference on Empirical Methods in Natural Language Processing (EMNLP)16 citationsJan 1, 2025Presumed Cultural Identity: How Names Shape LLM ResponsesHugging Face · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)16 citationsJan 1, 2025ThermalGen: Style-Disentangled Flow-Based Generative Models for RGB-to-Thermal Image TranslationTechnology Innovation Institute · neural information processing systems16 citationsJan 1, 2025Causal Head Gating: A Framework for Interpreting Roles of Attention Heads in TransformersMicrosoft · neural information processing systems16 citationsJan 1, 2025Asymmetric REINFORCE for off-Policy Reinforcement Learning: Balancing positive and negative rewardsGoogle DeepMind, Meta · neural information processing systems16 citationsJan 1, 2025MURI: High-Quality Instruction Tuning Datasets for Low-Resource Languages via Reverse InstructionsCohere · Transactions of the Association for Computational Linguistics16 citationsJan 1, 2025ReflectionCoder: Learning from Reflection Sequence for Enhanced One-off Code GenerationSenseTime · Proceedings of the conference - Association for Computational Linguistics. Meeting16 citationsJan 1, 2025Improving Contextual Faithfulness of Large Language Models via Retrieval Heads-Induced OptimizationiFlytek · Proceedings of the conference - Association for Computational Linguistics. Meeting16 citationsJan 1, 2025IOPO: Empowering LLMs with Complex Instruction Following via Input-Output Preference OptimizationAlibaba · Proceedings of the conference - Association for Computational Linguistics. Meeting16 citationsJan 1, 2025Exposing the Achilles’ Heel: Evaluating LLMs Ability to Handle Mistakes in Mathematical ReasoningMicrosoft · Proceedings of the conference - Association for Computational Linguistics. Meeting16 citationsJan 1, 2025Scaling Laws for Multilingual Language ModelsMicrosoft · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)16 citationsJan 1, 2025Failing Forward: Improving Generative Error Correction for ASR with Synthetic Data and Retrieval AugmentationMeta, Microsoft · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)16 citationsJan 1, 2025R3Mem: Bridging Memory Retention and Retrieval via Reversible CompressionAlphabet · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)16 citationsJan 1, 2025TRACT: Regression-Aware Fine-tuning Meets Chain-of-Thought Reasoning for LLM-as-a-JudgeAlphabet · Proceedings of the conference - Association for Computational Linguistics. Meeting16 citationsJan 1, 2025GenSim: A General Social Simulation Platform with Large Language Model based AgentsAlibaba · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)16 citationsJan 1, 2025SELFGOAL: Your Language Agents Already Know How to Achieve High-level GoalsAllen Institute for AI · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)16 citationsJan 1, 2025The emergence of sparse attention: impact of data distribution and benefits of repetitionGoogle DeepMind · neural information processing systems15 citationsJan 1, 2025Tiled Flash Linear Attention: More Efficient Linear RNN and xLSTM KernelsGoogle DeepMind · neural information processing systems15 citationsJan 1, 2025Generative Pre-trained Autoregressive Diffusion TransformerMicrosoft · neural information processing systems15 citationsJan 1, 2025ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented GenerationMicrosoft · neural information processing systems15 citationsJan 1, 2025Purifying Shampoo: Investigating Shampoo's Heuristics by Decomposing its PreconditionerMeta · neural information processing systems15 citationsJan 1, 2025DIPO: Dual-State Images Controlled Articulated Object Generation Powered by Diverse DataHorizon Robotics · neural information processing systems15 citationsJan 1, 2025MetaFaith: Faithful Natural Language Uncertainty Expression in LLMsAlphabet · Conference on Empirical Methods in Natural Language Processing (EMNLP)15 citationsJan 1, 2025Struct2D: A Perception-Guided Framework for Spatial Reasoning in MLLMsMicrosoft · neural information processing systems15 citationsJan 1, 2025