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

Aug 24

Submitted by

eric-xiang

Graph Engineering in the Era of LLM Agents: From Individual Intelligence to System Intelligence

·
35 authors

Submitted by

YerbaPage

ParaTempo: Efficient Parallel Reasoning via Temporal Confidence

SJTU

Shanghai Jiao Tong University

Submitted by

taesiri

InfinityEdit: Infinite Video Editing with a Lightweight Edit-Ignition Adapter

·
12 authors

Submitted by

zsytony

Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs

tencent

Submitted by

lana-ny

Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts

kakaocorp

3

Submitted by

BradyFU

OmniAssistBench: Assistant-style Interaction Benchmark for Omni-LLMs

nanjinguniv

Nanjing University

Submitted by

DeyangKong

Every Coin Has Two Sides: On the Dual Nature of Generalization in On-Policy Distillation of Large Language Models

IQuestLab

2

Submitted by

teawhite

WorldMind: Decoupled Game World Model for State-Aware NPC Behavior

·
4 authors

Submitted by

Minbyul

AgentMercury: Your Agent Can Synthesize Verifiable Environments for Business Scenarios at scale

·
2 authors

Submitted by

yjb6

UniSpace: Unified Visual Representation and Scalable Multimodal Modeling

meituan-longcat

Submitted by

EnjunDu

EviRank: Structured Relevance Evidence for Multimodal Image Re-ranking

·
11 authors

2

Submitted by

PardisTaghavi

Partition the Support, Reconstruct the Residual: Training-Free Sparse Attention for Video Generation and World Models

TexasAMUniversity

Texas A&M University

4

Submitted by

hongyu-nvidia

Hydra-0: Action Flow for Generalist World Modeling and Control

nvidia

2

Submitted by

taesiri

Llama-Mobile: Efficient 2.7-Bit Quantization of VLMs

·
3 authors

Submitted by

cseeychen

Human-Centric Intelligence in the Era of Foundation Models: A Survey

·
16 authors

Submitted by

Unseen1980

Daedalus-150M: A Convolution-Attention Hybrid Designed for CPU Inference

·
1 authors

Submitted by

TangJiakai5704

Towards Faithful Simulation of Human Shopping Behavior

·
12 authors

2

Submitted by

EnyiJiang

CLEAR: Continuous Latent Adapter Routing for Utility-Preserving LLM Safety Alignment

UIUC-CS

University of Illinois at Urbana-Champaign

Submitted by

cmhungsteve

PhysCaP: Grounding Code-as-Policy Agent with Physics-Informed Exploration

·
11 authors

2

Submitted by

ranausmans

Peer-Voted LLM-Agent Stress Tests Find Feed-Induced Lexical Convergence but No Reliable Matched-Exposure Advantage for Distributed Sources

·
2 authors

Submitted by

josefchen

FlavourBench: Ranking Frontier Language Models with Executable Culinary Ground Truth

·
2 authors

Submitted by

Minbyul

Hadith computational science in the age of large language models: a critical narrative review

·
2 authors