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

Oct 23

Submitted by

taesiri

Every Attention Matters: An Efficient Hybrid Architecture for Long-Context Reasoning

antgroup

4

Submitted by

WooooDyy

BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping

·
21 authors

Submitted by

lynazhang

LoongRL:Reinforcement Learning for Advanced Reasoning over Long Contexts

MicrosoftResearch

Microsoft Research

Submitted by

tmencatt

Language Models are Injective and Hence Invertible

gladia

Gladia Research Group

2

Submitted by

maximorulli

Attention Sinks in Diffusion Language Models

·
6 authors

2

Submitted by

Jeff-Wang

GigaBrain-0: A World Model-Powered Vision-Language-Action Model

open-gigaai

Submitted by

BK-Lee

Unified Reinforcement and Imitation Learning for Vision-Language Models

nvidia

4

Submitted by

GMFTBY

DeepWideSearch: Benchmarking Depth and Width in Agentic Information Seeking

AIDC-AI

Submitted by

GMFTBY

HSCodeComp: A Realistic and Expert-level Benchmark for Deep Search Agents in Hierarchical Rule Application

AIDC-AI

Submitted by

taesiri

Pico-Banana-400K: A Large-Scale Dataset for Text-Guided Image Editing

apple

Submitted by

taesiri

VideoAgentTrek: Computer Use Pretraining from Unlabeled Videos

·
15 authors

Submitted by

AIGCer-OPPO

DaMo: Data Mixing Optimizer in Fine-tuning Multimodal LLMs for Mobile Phone Agents

OPPOer

OPPO

Submitted by

ChaoHuangCS

Directional Reasoning Injection for Fine-Tuning MLLMs

·
10 authors

Submitted by

MorVentura

DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models

·
5 authors

Submitted by

taesiri

olmOCR 2: Unit Test Rewards for Document OCR

allenai

Ai2

Submitted by

js-hyun

Decomposed Attention Fusion in MLLMs for Training-Free Video Reasoning Segmentation

·
6 authors

Submitted by

taesiri

ColorAgent: Building A Robust, Personalized, and Interactive OS Agent

·
23 authors

Submitted by

kailinjiang

KORE: Enhancing Knowledge Injection for Large Multimodal Models via Knowledge-Oriented Augmentations and Constraints

·
10 authors

Submitted by

Arlolo0

OmniNWM: Omniscient Driving Navigation World Models

·
11 authors

Submitted by

jinjiajie

FinSight: Towards Real-World Financial Deep Research

·
6 authors

Submitted by

EunsuKim

Are they lovers or friends? Evaluating LLMs' Social Reasoning in English and Korean Dialogues

·
8 authors

Submitted by

hhenryz

From Charts to Code: A Hierarchical Benchmark for Multimodal Models

·
10 authors

Submitted by

rezaesfandiarpoor

TheMCPCompany: Creating General-purpose Agents with Task-specific Tools

·
5 authors

Submitted by

ZacharyNovack

Steering Autoregressive Music Generation with Recursive Feature Machines

·
5 authors

2

Submitted by

shizhediao

ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge

nvidia

Submitted by

ZhiZhang

NeuroAda: Activating Each Neuron's Potential for Parameter-Efficient Fine-Tuning

uva

University of Amsterdam

Submitted by

kailinjiang

MINED: Probing and Updating with Multimodal Time-Sensitive Knowledge for Large Multimodal Models

·
15 authors

1

Submitted by

mandipgoswami

RIR-Mega: a large-scale simulated room impulse response dataset for machine learning and room acoustics modeling

·
1 authors

Submitted by

ling441

See the Text: From Tokenization to Visual Reading

·
6 authors

2

Submitted by

quao627

AlphaOPT: Formulating Optimization Programs with Self-Improving LLM Experience Library

·
13 authors

Submitted by

lx865712528

Learning from the Best, Differently: A Diversity-Driven Rethinking on Data Selection

·
9 authors

Submitted by

nielsr

Accelerating Vision Transformers with Adaptive Patch Sizes

Carnegie Mellon University School of Computer Science

Submitted by

deqing

When Do Transformers Learn Heuristics for Graph Connectivity?

·
4 authors

2

Submitted by

yanhong-l

Text or Pixels? It Takes Half: On the Token Efficiency of Visual Text Inputs in Multimodal LLMs

·
3 authors

Submitted by

lwachowiak

What Questions Should Robots Be Able to Answer? A Dataset of User Questions for Explainable Robotics

·
4 authors

Submitted by

liamdugan

Machine Text Detectors are Membership Inference Attacks

·
5 authors

Submitted by

Yuan-avs

SAVANT: Semantic Analysis with Vision-Augmented Anomaly deTection

·
5 authors

2