Profiles2021-2026
Qiang Liu
New Laboratory of Pattern Recognition (NLPR), Institute of Automation, Chinese Academy of Sciences (CASIA)
44Papers
0Last author
61Unique co-authors
1First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers44
- AVoCaDO: An Audiovisual Video Captioner Driven by Temporal OrchestrationAuthor 9 of 12ICLR 2026 · 27 citations
- OmniSIFT: Modality-Asymmetric Token Compression for Efficient Omni-modal Large Language ModelsAuthor 13 of 15ICML 2026 · 22 citations
- VidBridge-R1: Bridging QA and Captioning for RL-based Video Understanding Models with Intermediate Proxy TasksAuthor 9 of 12ICLR 2026 · 7 citations
- Disentangling Knowledge Representations for Large Language Model EditingAuthor 4 of 7ICLR 2026 · 6 citations
- Uncovering Competing Poisoning Attacks in Retrieval-Augmented GenerationAuthor 6 of 9KDD 2026 · 4 citations
- SpectraLLM: Uncovering the Ability of LLMs for Molecule Structure Elucidation from Multi-SpectraAuthor 6 of 7ICLR 2026 · 1 citation
- CoT is Not the Chain of Truth: An Empirical Internal Analysis of Reasoning LLMs for Fake News GenerationAuthor 5 of 8ICML 2026
- Reading the Cell, Designing the Cure: Perturbation-Conditioned Molecular Diffusion for Function-Oriented Drug DesignAuthor 5 of 7ICML 2026
- RLShield: Dynamic Jailbreak Detection for LLMs via Reinforced Adaptive LearningAuthor 5 of 8ACL Findings 2026
- Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual DrawingAuthor 4 of 8NeurIPS 2025 · 153 citations
- Personalized Text Generation with Contrastive Activation SteeringAuthor 4 of 7ACL 2025 · 24 citations
- CoRA: Collaborative Information Perception by Large Language Model's Weights for RecommendationAuthor 5 of 8AAAI 2025 · 15 citations
- Generate First, Then Sample: Enhancing Fake News Detection with LLM-Augmented Reinforced SamplingAuthor 4 of 7ACL 2025 · 14 citations
- Divide-Then-Align: Honest Alignment based on the Knowledge Boundary of RAGAuthor 4 of 12ACL 2025 · 11 citations
- Fairness without Demographics through Learning Graph of GradientsAuthor 3 of 5KDD 2025 · 3 citations
- Attention-guided Self-reflection for Zero-shot Hallucination Detection in Large Language ModelsAuthor 1 of 7EMNLP 2025 · 2 citations
- GenPilot: A Multi-Agent System for Test-Time Prompt Optimization in Image GenerationAuthor 8 of 9EMNLP Findings 2025 · 2 citations
- SINCon: Mitigate LLM-Generated Malicious Message Injection Attack for Rumor DetectionAuthor 2 of 5ACL 2025 · 2 citations
- VidCapBench: A Comprehensive Benchmark of Video Captioning for Controllable Text-to-Video GenerationAuthor 8 of 10ACL Findings 2025 · 2 citations
- SHARP: Steering Hallucination in LVLMs via Representation EngineeringAuthor 7 of 10EMNLP 2025
- Toolscaler: Scalable Generative Tool Calling via Structure-Aware Semantic TokenizationAuthor 8 of 9EMNLP Findings 2025
- Uncovering Overfitting in Large Language Model EditingAuthor 3 of 6ICLR 2025
- Text-Guided Molecule Generation with Diffusion Language ModelAuthor 2 of 4AAAI 2024 · 45 citations
- Rethinking Graph Masked Autoencoders through Alignment and UniformityAuthor 3 of 4AAAI 2024 · 40 citations
- Chain-of-History Reasoning for Temporal Knowledge Graph ForecastingAuthor 3 of 6ACL Findings 2024 · 35 citations
- Logical Closed Loop: Uncovering Object Hallucinations in Large Vision-Language ModelsAuthor 2 of 7ACL Findings 2024 · 24 citations
- Modality-Balanced Learning for Multimedia RecommendationAuthor 3 of 6ACM MM 2024 · 21 citations
- Heterogeneous Graph Reasoning for Fact Checking over Texts and TablesAuthor 4 of 5AAAI 2024 · 19 citations
- Rethinking Fair Graph Neural Networks from Re-balancingAuthor 3 of 4KDD 2024 · 12 citations
- Pin-Tuning: Parameter-Efficient In-Context Tuning for Few-Shot Molecular Property PredictionAuthor 2 of 5NeurIPS 2024 · 10 citations
Showing the 30 most recent indexed papers.
