Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
DMC-CF: Dynamic Multimodal CounterFactual QA benchmark for Causal ReasoningAlibaba · arXiv (Cornell University)May 28Robust and Generalizable Safety Steering for Text-to-Image Diffusion TransformersAlibaba · arXiv (Cornell University)May 28Make LLM Learn to Synthesize from Streaming Experiences through FeedbackAlibaba · arXiv (Cornell University)May 28On-Policy Replay for Continual Supervised Fine-TuningAlibaba · arXiv (Cornell University)May 28FinGuard: Detecting Financial Regulatory Non-Compliance in LLM InteractionsAlibaba · arXiv (Cornell University)May 28DP-SAPF: Saliency-Aware Parameter Fine-tuning of Public Models for Differentially Private Image SynthesisMicrosoft · arXiv (Cornell University)May 28RTP-LLM: High-Performance Alibaba LLM Inference EngineAlibaba · arXiv (Cornell University)May 28How Coding Agents Fail Their Users: A Large-Scale Analysis of Developer-Agent Misalignment in 20,574 Real-World SessionsAlphabet · arXiv (Cornell University)May 28LFQ: Logit-aware Final-block Quantization for Boosting the Generation Quality of Low-Bit Quantized LLMsLG AI Research · arXiv (Cornell University)May 28EvoRubric: Self-Evolving Rubric-Driven RL for Open-Ended GenerationAlibaba · arXiv (Cornell University)May 28Why Larger Models Learn More: Effects of Capacity, Interference, and Rare-Task RetentionAnthropic · arXiv (Cornell University)May 28Evolving Skill-Structured Attack Memory Enhances LLM JailbreakingAlibaba · arXiv (Cornell University)May 28On Language Generation in the Limit with Bounded MemoryAlphabet · arXiv (Cornell University)May 28STAMP: Training Explicit Memory for Mobile GUI Agents in Controllable and Scalable Virtual EnvironmentsAlibaba · arXiv (Cornell University)May 28Decoding Strategies for Diffusion-Based ASR: A Systematic Evaluation of Confidence-Based ThresholdingAlphabet · arXiv (Cornell University)May 28GDSD: Reinforcement Learning as Guided Denoiser Self-Distillation for Diffusion Language ModelsAlibaba · arXiv (Cornell University)May 28MELD: Mel-Spectrogram-Based Speech Language Modeling with Discrete Latent VariablesGoogle DeepMind · arXiv (Cornell University)May 28Rethinking Post-Training Recipes for Multimodal Time-Series ForecastingAlphabet · arXiv (Cornell University)May 28GRASP: Gated Regression-Aware Skill Proposer for Self-Improving LLM AgentsMicrosoft · arXiv (Cornell University)May 28ESAM++: Efficient Online 3D Perception on the EdgeAlphabet · arXiv (Cornell University)May 28Test Time Training for Supervised Causal LearningMicrosoft · arXiv (Cornell University)May 28PEARL: Training Socratic Tutors with Pedagogically Aligned Reinforcement LearningiFlytek · arXiv (Cornell University)May 28CrystalXRD-Bench: Benchmarking Vision-Language Models for XRD Peak Indexing Across Diverse Crystalline MaterialsAlibaba · arXiv (Cornell University)May 28Demystifying Data Organization for Enhanced LLM TrainingMicrosoft · arXiv (Cornell University)May 28Evaluating Cross-lingual Knowledge Consistency in Code-Mixed vis-a-vis Indian Languages using IndicKLARMicrosoft · arXiv (Cornell University)May 28