📄 ACL 2024: RGCL, Retrieval-Guided Contrastive Learning for Hateful Meme Detection 📄 EMNLP 2025 (Oral): RA-HMD, Robust Adaptation of Large Multimodal Models for Retrieval-Augmented Hateful Meme Detection Official implementation with pretrained models and reproduction scripts.

ESGenius: EMNLP 2025 Main Oral benchmark for LLM ESG and sustainability knowledge, with 1,136 questions, evaluation code, interactive heatmap, and Hugging Face dataset release.

Official implementation of "X-CoT: Explainable Text-to-Video Retrieval via LLM-based Chain-of-Thought Reasoning (EMNLP 2025)"

EMNLP 2025 Findings | SelfAug: distribution self-alignment to mitigate forgetting in RAG fine-tuning.

Official implementation of "CSKS: Continuously Steering LLMs Sensitivity to Contextual Knowledge with Proxy Models" (EMNLP 2025)

ESGenius: EMNLP 2025 Main Oral benchmark for LLM ESG and sustainability knowledge, with 1,136 questions, evaluation code, interactive heatmap, and Hugging Face dataset release.

Companion repository for A Survey of Cognitive Distortion Detection and Classification in NLP (EMNLP 2025 Findings)

Replication of 'From Reasoning to Answer' (EMNLP 2025) — Reasoning-Focus Heads + Activation Patching on DeepSeek-R1-Distill-Qwen-7B