Pheng-Ann Heng

Lost in Tokenization: Context as the Key to Unlocking Biomolecular Understanding in Scientific LLMs

EgoHandICL: Egocentric 3D Hand Reconstruction with In-Context Learning

reAR: Rethinking Visual Autoregressive Models via Token-wise Consistency Regularization

Omni-Captioner: Data Pipeline, Models, and Benchmark for Omni Detailed Perception

DEL-Ranking: Ranking-Correction Denoising Framework for Elucidating Molecular Affinities in DNA-Encoded Libraries

ProteinAE: Protein Diffusion Autoencoders for Structure Encoding

MME-Emotion: A Holistic Evaluation Benchmark for Emotional Intelligence in Multimodal Large Language Models

PathFinder: Graph-structured Reasoning for Medical Visual Question Answering

Lightweight MSA Design Advances Protein Folding From Evolutionary Embeddings

SurgRe4DGS: Open-Vocabulary Vision-Language Gaussian Splatting for Retrievable 4D Surgical Scene Reconstruction

KnowGuard: Knowledge-Driven Abstention for Multi-Round Clinical Reasoning

RiboPO: Preference Optimization for Structure- and Stability-Aware RNA Design

HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model

GPattern-Bench: Benchmarking Gene Spatial Pattern Classification in Subcellular Spatial Transcriptomics

CA-DEL: Construction and Application of an Intelligent DEL Database for Anti-Cancer Drug Discovery

Boosting Multiagent Reinforcement Learning at High Replay Ratios with Ensemble Reset

InstructPLM-mu: 1-Hour Fine-Tuning of ESM2 Beats ESM3 in Protein Mutation Predictions

From Supervision to Exploration: What Does Protein Language Model Learn During Reinforcement Learning?

ScaleAC: Scale Actor-Critic by Replay Ratio

DisCo-Layout: Disentangling and Coordinating Semantic and Physical Refinement in a Multi-Agent Framework for 3D Indoor Layout Synthesis

Boosting In-Silicon Directed Evolution with Fine-Tuned Protein Language Model and Tree Search

RAGCell: Retrieval-Augmented Generation as Supervision for Versatile Single-cell Analysis

Delving into RL for Image Generation with CoT: A Study on DPO vs. GRPO

GLID$^2$E: A Gradient-Free Lightweight Fine-tune Approach for Discrete Biological Sequence Design

T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT

SceneDecorator: Towards Scene-Oriented Story Generation with Scene Planning and Scene Consistency

Fast-in-Slow: A Dual-System VLA Model Unifying Fast Manipulation within Slow Reasoning

COS3D: Collaborative Open-Vocabulary 3D Segmentation

Protein Inverse Folding From Structure Feedback

What We Miss Matters: Learning from the Overlooked in Point Cloud Transformers

HybridVLA: Collaborative Autoregression and Diffusion in a Unified Vision-Language-Action Model

Object-aware lifting for 3D scene segmentation in Gaussian splatting

PointACL: Point Cloud Understanding via Attention-Driven Contrastive Learning

MagicTailor: Component-Controllable Personalization in Text-to-Image Diffusion Models

SeaDAG: Semi-autoregressive Diffusion for Conditional Directed Acyclic Graph Generation

DEL-Ranking: Ranking-Correction Denoising Framework for Elucidating Molecular Affinities in DNA-Encoded Libraries