AI.info

Latest Research — Page 17

Browse Latest Research on AI.info.

• CombiGraph-Vis: A Curated Multimodal Olympiad Benchmark for Discrete Mathematical Reasoning
• Prime Agent: A Self-Improving RLM Harness
• On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability
• Large Language Model Reasoning Failures
• Approximating $f$-Divergences with Rank Statistics
• Scaling Lean Theorem Proving for Real-World Code Verification
• PPA-Plan: Proactive Pitfall Avoidance for Reliable Planning in Long-Context LLM Reasoning
• ORLoopBench: Solver-in-the-Loop Benchmarks for Self-Correction and Behavioral Rationality in Operations Research
• Sensor-Language-Action Models
• 4D-RaDiff: Latent Point Diffusion for 4D Radar Point Cloud Generation
• Context-weighted Discrete Flow Matching
• Annotation-Efficient Universal Honesty Alignment
• Alterbute: Editing Intrinsic Attributes of Objects in Images
• BaRe-Mem: Bayesian Reliability Memory for Robust and Adaptive Agent Consultation
• VibeVoice-ASR-Streaming Technical Report
• Encoder-Decoder Diffusion Language Models for Efficient Training and Inference
• Self-Credit: Credit Assignment for Long-Horizon Agentic RL via Intra-group Rescaling
• Differentiable Logic Gate Networks for Low-Latency EEG Classification on Edge Devices
• CRESiST: Self-Enhancing Explorative Policy Learning for Simultaneous Speech Translation
• NeoMME: A Single-Tower Multimodal-Native Multilingual Foundation Encoder for Efficient Fine-Tuning and Inference
• CARVE: Verified Expansion for Variable-Length Generation in Diffusion Language Models
• Enginuity: Building an Open Multi-Domain Dataset of Complex Engineering Diagrams
• Chain-of-Thought as a Lens: Evaluating Structured Reasoning Alignment between Human Preferences and Large Language Models
• Tell Me: An LLM-powered Mental Well-being Assistant with RAG, Synthetic Dialogue Generation, and Agentic Planning
• Nested Sparsity for Scalable Pretraining and Continual Learning
• A Human-in-the-Loop Corpus for LLM-Based Simplification of Scientific Summaries
• Enhancing Pre-trained Representation Classifiability can Boost its Interpretability
• Report for NSF Workshop on AI for Electronic Design Automation
• MetaShare: Failure-Aware Collective Self-Improvement for Open-Ended Agents
• SPASM: Stable Persona-driven Agent Simulation for Multi-turn Dialogue Generation
• Unlocking Lossless Speedups in LLMs via Discrete Diffusion
• Surface Conditioning for Zero-Shot Multivariate Correction in a Marine Biogeochemistry Emulator
• Robustness study of the bio-inspired musculoskeletal arm robot based on the data-driven iterative learning algorithm
• Efficient Zero-Shot Inpainting with Decoupled Diffusion Guidance
• Broken on Arrival: Non-Functional Quantized LLM Artifacts in Public Model Registries and How to Catch Them
• OMG-Bench: A New Challenging Benchmark for Skeleton-based Online Micro Hand Gesture Recognition
• UserArena: Benchmarking Interactive User Simulation for Long-Horizon Shopping Agents
• ASAP: an Agentic Solution to Auto-optimize Performance of Large-Scale LLM Training
• The Sharp Directions Are Against You: Curvature Analysis of Activation Steering
• CREward: A Type-Specific Creativity Reward Model