AI Research Papers Daily

Crowdsource, Crawl, or Generate? Creating SEA-VL, a Multicultural Vision-Language Dataset for Southeast Asia

LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RL

YuE: Scaling Open Foundation Models for Long-Form Music Generation

Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuning

MagicInfinite: Generating Infinite Talking Videos with Your Words and Voice

Seedream 2.0: A Native Chinese-English Bilingual Image Generation Foundation Model

Video Action Differencing

UniF^2ace: Fine-grained Face Understanding and Generation with Unified Multimodal Models

Gemini Embedding: Generalizable Embeddings from Gemini

SegAgent: Exploring Pixel Understanding Capabilities in MLLMs by Imitating Human Annotator Trajectories

Tuning-Free Multi-Event Long Video Generation via Synchronized Coupled Sampling

LightGen: Efficient Image Generation through Knowledge Distillation and Direct Preference Optimization

Implicit Reasoning in Transformers is Reasoning through Shortcuts

Exploiting Instruction-Following Retrievers for Malicious Information Retrieval

OmniMamba: Efficient and Unified Multimodal Understanding and Generation via State Space Models

CineBrain: A Large-Scale Multi-Modal Brain Dataset During Naturalistic Audiovisual Narrative Processing

"Principal Components" Enable A New Language of Images

Robusto-1 Dataset: Comparing Humans and VLMs on real out-of-distribution Autonomous Driving VQA from Peru

VisualSimpleQA: A Benchmark for Decoupled Evaluation of Large Vision-Language Models in Fact-Seeking Question Answering

^RFLAV: Rolling Flow matching for infinite Audio Video generation

Benchmarking AI Models in Software Engineering: A Review, Search Tool, and Enhancement Protocol

Mixture of Experts Made Intrinsically Interpretable

BiasEdit: Debiasing Stereotyped Language Models via Model Editing

AnyMoLe: Any Character Motion In-betweening Leveraging Video Diffusion Models

AI-native Memory 2.0: Second Me

Inductive Moment Matching

LocAgent: Graph-Guided LLM Agents for Code Localization

Perplexity Trap: PLM-Based Retrievers Overrate Low Perplexity Documents

Referring to Any Person

RayFlow: Instance-Aware Diffusion Acceleration via Adaptive Flow Trajectories

QuoTA: Query-oriented Token Assignment via CoT Query Decouple for Long Video Comprehension

Beyond Decoder-only: Large Language Models Can be Good Encoders for Machine Translation

Collapse of Dense Retrievers: Short, Early, and Literal Biases Outranking Factual Evidence

Evaluating Intelligence via Trial and Error

ObjectMover: Generative Object Movement with Video Prior

Capacity-Aware Inference: Mitigating the Straggler Effect in Mixture of Experts

NullFace: Training-Free Localized Face Anonymization

Ideas in Inference-time Scaling can Benefit Generative Pre-training Algorithms

PlainQAFact: Automatic Factuality Evaluation Metric for Biomedical Plain Language Summaries Generation

OTTER: A Vision-Language-Action Model with Text-Aware Visual Feature Extraction