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

Jul 15

Submitted by

ubowang

Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models

TIGER-Lab

Submitted by

huangrh9

Read It Back: Pretrained MLLMs Are Zero-Shot Reward Models for Text-to-Image Generation

ByteDance-Seed

Submitted by

CongWei1230

Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation

TIGER-Lab

Submitted by

amitsaha

SynthDocBench: Controlled Benchmark for Long-Context Visual Document Understanding

ServiceNow-AI

Submitted by

chengkunli

Blind-Spots-Bench: Evaluating Blind Spots in Multimodal Models

epfl-ch

Submitted by

Saint-lsy

Towards Autonomous and Auditable Medical Imaging Model Development

CUOHK

Chinese University of Hong Kong

Submitted by

nielsr

MonkeyOCRv2: A Visual-Text Foundation Model for Document AI

Submitted by

ameroyer

MuScriptor: An Open Model for Multi-Instrument Music Transcription

kyutai

Submitted by

Yushi98

Navigating the Mirage: A Dual-Path Agentic Framework for Robust Misleading Chart Question Answering

Submitted by

Silin-Chen

Know Before Fix: QA-Driven Repository Knowledge Acquisition for Software Issue Resolution

SJTU

Shanghai Jiao Tong University

Submitted by

taesiri

Let RGB Be the Language of Vision

·
14 authors

Submitted by

ezgikorkmaz

Principled Analysis of Deep Reinforcement Learning Evaluation and Design Paradigms

·
1 authors

Submitted by

Yushi98

Are LLMs Ready for Scientific Discovery? A Capability-Oriented Benchmark for AI Scientists

Submitted by

pratyushrt

What LLM Forecasters Know but Don't Say: Probing Internal Representations for Calibration and Faithfulness

eternisai

Submitted by

iforgott

MAGIC: Transition-Aware Generation of Navigable Multi-Scene Game Worlds with Large Language Models

CUHK-CSE

The Chinese University of Hong Kong