new

Get trending papers in your email inbox once a day!

Get trending papers in your email inbox!

Daily Papers

byAK and the research community

Jan 20

Submitted by

red-fox-yj

ABC-Bench: Benchmarking Agentic Backend Coding in Real-World Development

OpenMOSS-Team

Submitted by

unilm

Multiplex Thinking: Reasoning via Token-wise Branch-and-Merge

MicrosoftResearch

Microsoft Research

Submitted by

JiayuJeff

NAACL: Noise-AwAre Verbal Confidence Calibration for LLMs in RAG Systems

·
10 authors

Submitted by

ChongCong

Medical SAM3: A Foundation Model for Universal Prompt-Driven Medical Image Segmentation

·
10 authors

Submitted by

taesiri

The Assistant Axis: Situating and Stabilizing the Default Persona of Language Models

·
5 authors

Submitted by

hheiden

PubMed-OCR: PMC Open Access OCR Annotations

bevaya

3

Submitted by

BiaoGong

CoDance: An Unbind-Rebind Paradigm for Robust Multi-Subject Animation

·
8 authors

Submitted by

rzdiversity

Spurious Rewards Paradox: Mechanistically Understanding How RLVR Activates Memorization Shortcuts in LLMs

·
8 authors

Submitted by

BounharAbdelaziz

YaPO: Learnable Sparse Activation Steering Vectors for Domain Adaptation

MBZUAI-Paris

MBZUAI-IFM Paris Lab

Submitted by

wanng

SIN-Bench: Tracing Native Evidence Chains in Long-Context Multimodal Scientific Interleaved Literature

IIGroup

Tsinghua IIGroup

Submitted by

ralfroemer

CLARE: Continual Learning for Vision-Language-Action Models via Autonomous Adapter Routing and Expansion

TUM

Technical University of Munich