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

Sep 26

Submitted by

Nothing2Say

VCRL: Variance-based Curriculum Reinforcement Learning for Large Language Models

·
7 authors

2

Submitted by

Sicong

MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources

·
15 authors

Submitted by

taesiri

SciReasoner: Laying the Scientific Reasoning Ground Across Disciplines

·
32 authors

Submitted by

xiaochonglinghu

Tree Search for LLM Agent Reinforcement Learning

·
6 authors

Submitted by

wujie10

Seedream 4.0: Toward Next-generation Multimodal Image Generation

·
50 authors

6

Submitted by

taesiri

Hunyuan3D-Omni: A Unified Framework for Controllable Generation of 3D Assets

·
19 authors

Submitted by

Samoed

AutoIntent: AutoML for Text Classification

·
4 authors

Submitted by

qianlanwyd

TrustJudge: Inconsistencies of LLM-as-a-Judge and How to Alleviate Them

·
14 authors

2

Submitted by

Suu

CE-GPPO: Controlling Entropy via Gradient-Preserving Clipping Policy Optimization in Reinforcement Learning

·
8 authors

Submitted by

intfloat

Thinking Augmented Pre-training

·
5 authors

2

Submitted by

ankile

Residual Off-Policy RL for Finetuning Behavior Cloning Policies

·
6 authors

Submitted by

chengle

Recon-Act: A Self-Evolving Multi-Agent Browser-Use System via Web Reconnaissance, Tool Generation, and Task Execution

·
4 authors

Submitted by

hyz317

CHARM: Control-point-based 3D Anime Hairstyle Auto-Regressive Modeling

·
9 authors

Submitted by

Shilin-LU

Does FLUX Already Know How to Perform Physically Plausible Image Composition?

·
6 authors

2

Submitted by

MingLiiii

Understanding the Thinking Process of Reasoning Models: A Perspective from Schoenfeld's Episode Theory

·
9 authors

2

Submitted by

taesiri

SD3.5-Flash: Distribution-Guided Distillation of Generative Flows

stabilityai

Submitted by

QizhiPei

ScaleDiff: Scaling Difficult Problems for Advanced Mathematical Reasoning

·
9 authors

Submitted by

CSJianYang

V-GameGym: Visual Game Generation for Code Large Language Models

·
12 authors

Submitted by

HaotongQin

Quantized Visual Geometry Grounded Transformer

·
11 authors

2

Submitted by

chengq9

UserRL: Training Interactive User-Centric Agent via Reinforcement Learning

·
13 authors

Submitted by

augustinLib

BESPOKE: Benchmark for Search-Augmented Large Language Model Personalization via Diagnostic Feedback

·
4 authors

Submitted by

TangJiakai5704

Interactive Recommendation Agent with Active User Commands

·
15 authors

2

Submitted by

penfever

When Judgment Becomes Noise: How Design Failures in LLM Judge Benchmarks Silently Undermine Validity

·
5 authors

3

Submitted by

taesiri

SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent

·
4 authors

Submitted by

Jungang

MOSS-ChatV: Reinforcement Learning with Process Reasoning Reward for Video Temporal Reasoning

·
11 authors

Submitted by

lx865712528

Behind RoPE: How Does Causal Mask Encode Positional Information?

·
6 authors

2

Submitted by

taesiri

StyleBench: Evaluating thinking styles in Large Language Models

·
5 authors

Submitted by

pengxiang

Discrete Diffusion for Reflective Vision-Language-Action Models in Autonomous Driving

·
9 authors

Submitted by

prateekv

Thinking While Listening: Simple Test Time Scaling For Audio Classification

·
2 authors

Submitted by

gberton

CompLLM: Compression for Long Context Q&A

amazon

2

Submitted by

TianheWu

The Unanticipated Asymmetry Between Perceptual Optimization and Assessment

·
5 authors

Submitted by

dlion168

MI-Fuse: Label Fusion for Unsupervised Domain Adaptation with Closed-Source Large-Audio Language Model

·
3 authors

2

Submitted by

zx1239856

OverLayBench: A Benchmark for Layout-to-Image Generation with Dense Overlaps

·
10 authors

Submitted by

huzaifas-sidhpurwala

Blueprints of Trust: AI System Cards for End to End Transparency and Governance

·
5 authors

Submitted by

jpatel0057

Evaluating Large Language Models for Detecting Antisemitism

iDRAMALab