AI研究论文每日精选

Is That Your Final Answer? Test-Time Scaling Improves Selective Question Answering

Soundwave: Less is More for Speech-Text Alignment in LLMs

Cramming 1568 Tokens into a Single Vector and Back Again: Exploring the Limits of Embedding Space Capacity

Magma: A Foundation Model for Multimodal AI Agents

Continuous Diffusion Model for Language Modeling

Phantom: Subject-consistent video generation via cross-modal alignment

Multimodal Mamba: Decoder-only Multimodal State Space Model via Quadratic to Linear Distillation

Rethinking Diverse Human Preference Learning through Principal Component Analysis

You Do Not Fully Utilize Transformer's Representation Capacity

FLAG-Trader: Fusion LLM-Agent with Gradient-based Reinforcement Learning for Financial Trading

SoFar: Language-Grounded Orientation Bridges Spatial Reasoning and Object Manipulation

SafeRoute: Adaptive Model Selection for Efficient and Accurate Safety Guardrails in Large Language Models

Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities?

OctoTools: An Agentic Framework with Extensible Tools for Complex Reasoning

PAFT: Prompt-Agnostic Fine-Tuning

RealSyn: An Effective and Scalable Multimodal Interleaved Document Transformation Paradigm

Atom of Thoughts for Markov LLM Test-Time Scaling

Text2World: Benchmarking Large Language Models for Symbolic World Model Generation

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

HeadInfer: Memory-Efficient LLM Inference by Head-wise Offloading

YOLOv12: Attention-Centric Real-Time Object Detectors

HealthGPT: A Medical Large Vision-Language Model for Unifying Comprehension and Generation via Heterogeneous Knowledge Adaptation

Eager Updates For Overlapped Communication and Computation in DiLoCo

Flow-of-Options: Diversified and Improved LLM Reasoning by Thinking Through Options

The Hidden Risks of Large Reasoning Models: A Safety Assessment of R1

Crowd Comparative Reasoning: Unlocking Comprehensive Evaluations for LLM-as-a-Judge

Pre-training Auto-regressive Robotic Models with 4D Representations

Injecting Domain-Specific Knowledge into Large Language Models: A Comprehensive Survey

FinMTEB: Finance Massive Text Embedding Benchmark

Perovskite-LLM: Knowledge-Enhanced Large Language Models for Perovskite Solar Cell Research

Scaling Autonomous Agents via Automatic Reward Modeling And Planning

Multilingual Encoder Knows more than You Realize: Shared Weights Pretraining for Extremely Low-Resource Languages

Harnessing Vision Models for Time Series Analysis: A Survey