Publications — Yiqun Sun
Research Scientist, Magellan Technology Research Institute (MTRI), Singapore. Each title links to a page with the abstract and full-text PDF. Each first-author repository includes a short research-context page (claims the paper supports, placement in the literature, limitations) and a CITATION.cff/CITATION.bib. Venue labels follow the CORE 2023 and CCF 2026 rankings.
Published papers
-
GGSS: Geodesic-Gated Spherical Steering for Inference-Time Debiasing of Generative Vision-Language Models
Conference on Empirical Methods in Natural Language Processing (EMNLP) Main Conference, 2026 CORE A* CCF B · arXiv 2608.25375
-
When Models Hear What They Expect: Diagnosing Prosodic Heuristics in Multimodal Sarcasm Detection
Findings of the Association for Computational Linguistics: EMNLP, 2026 CORE A* CCF B · arXiv 2608.30204
-
Very Efficient Listwise Multimodal Reranking for Long Documents
International Conference on Machine Learning (ICML), 2026 CORE A* CCF A · arXiv 2605.11864
-
Corpus2Skill: Distilling Enterprise Knowledge into Navigable Agent Skills for QA and RAG
Findings of the Association for Computational Linguistics: EMNLP, 2026 CORE A* CCF B · arXiv 2604.14572
-
Position: Text Embeddings Should Capture Implicit Semantics, Not Just Surface Meaning
International Conference on Machine Learning (ICML), Position Paper Track, 2026 CORE A* CCF A · arXiv 2506.08354
-
Return of Frustratingly Easy Unsupervised Video Domain Adaptation
International Conference on Machine Learning (ICML), 2026 CORE A* CCF A · arXiv 2605.19510
-
DeepConnect: A Visual Analytics System for Bridging Interdisciplinary Research Collaborations
IEEE Transactions on Visualization and Computer Graphics (TVCG), 2026 CCF A · arXiv 2608.05134
-
Balancing Relevance and Diversity in k-Maximum Inner Product Search
The VLDB Journal, 35(4), 2026 CCF A · DOI 10.1007/s00778-026-00982-8 · arXiv 2402.13858
-
Partially Shared Concept Bottleneck Models
AAAI Conference on Artificial Intelligence (AAAI), 2026 CORE A* CCF A · arXiv 2511.22170
-
PRISM: A Framework for Producing Interpretable Political Bias Embeddings with Political-Aware Cross-Encoder
Annual Meeting of the Association for Computational Linguistics (ACL) Main Conference (Oral), 2025 CORE A* CCF A · DOI 10.18653/v1/2025.acl-long.1344 · arXiv 2505.24646
-
Don't Reinvent the Wheel: Efficient Instruction-Following Text Embedding based on Guided Space Transformation
Annual Meeting of the Association for Computational Linguistics (ACL) Main Conference (Oral), 2025 CORE A* CCF A · DOI 10.18653/v1/2025.acl-long.1196 · arXiv 2505.24754
-
A General Framework for Producing Interpretable Semantic Text Embeddings
International Conference on Learning Representations (ICLR), 2025 CORE A* CCF A · arXiv 2410.03435
-
Uncovering the Bigger Picture: Comprehensive Event Understanding Via Diverse News Retrieval
Conference on Empirical Methods in Natural Language Processing (EMNLP) Main Conference (Oral), 2025 CORE A* CCF B · arXiv 2508.19758
-
DiversiNews: Enriching News Consumption with Relevant Yet Diverse News Articles Retrieval
Proceedings of the VLDB Endowment (PVLDB) 17(12), Demo Track, 2024 CORE A* CCF A · DOI 10.14778/3685800.3685854
Preprints
-
Information Retrieval Misses the Mark for LLM Agents
SSRN preprint (position paper), 2026 · DOI 10.2139/ssrn.6903579
-
Sparse Autoencoders as Plug-and-Play Firewalls for Adversarial Attack Detection in VLMs
arXiv preprint, 2026 · arXiv 2605.07447
-
OrchMAS: Orchestrated Reasoning with Multi Collaborative Heterogeneous Scientific Expert Structured Agents
arXiv preprint, 2026 · arXiv 2603.03005
-
One Swallow Does Not Make a Summer: Understanding Semantic Structures in Embedding Spaces
arXiv preprint, 2025 · arXiv 2512.00852
Generated from src/data/publications.json at build time.