Papers by Songlin Wang

14 papers
LyapLock: Bounded Knowledge Preservation in Sequential Large Language Model Editing (2025.emnlp-main)

Copied to clipboard

Challenge: Existing models for enhancing knowledge updating are prone to performance degradation due to incomplete knowledge preservation mechanisms.
Approach: They propose a model for locate-then-edit that decomposes long-term constrained programming into tractable stepwise subproblems for efficient solving.
Outcome: The proposed framework achieves asymptotic optimal editing performance while meeting the constraints of long-term knowledge preservation.
More Thinking, Less Talking: Internalizing Deliberative Safety into LLM Parameters (2026.acl-long)

Copied to clipboard

Challenge: Existing safety alignment methods leave Large Language Models vulnerable to sophisticated jailbreak attacks.
Approach: They propose a safety reasoning internalization framework that internalizes safety reasoning into an implicit computational pathway using Low-Rank Adaptation (LoRA).
Outcome: The proposed framework achieves a 43% lower Attack Success Rate (ASR) against distinct jailbreak attacks compared to strong baselines.
Can Large Language Models Understand DL-Lite Ontologies? An Empirical Study (2024.findings-emnlp)

Copied to clipboard

Challenge: Large language models (LLMs) have shown remarkable proficiency in understanding textual data and revolutionizing the field of natural language processing.
Approach: They empirically analyze LLMs' capability of understanding Description Logic (DL) ontologies covering 6 representative tasks from syntactic and semantic aspects.
Outcome: The proposed model can understand formal syntax and model-theoretic semantics of concepts and roles, but struggle with understanding TBox NI transitivity and handling ontologies with large ABoxes.
Gamma-Guard: Lightweight Residual Adapters for Robust Guardrails in Large Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) are widely deployed as zero-shot evaluators for answer grading, content moderation, and document ranking.
Approach: They propose a system that trains LLMs with adapters to denoise embeddings and refocus attention.
Outcome: The proposed model lifts adversarial accuracy from 5% to 95% a 90 percentage-point gain while reducing clean-data accuracy by just 8 percentage points.
ESimCSE: Enhanced Sample Building Method for Contrastive Learning of Unsupervised Sentence Embedding (2022.coling-1)

Copied to clipboard

Challenge: a new method for learning unsupervised sentence embeddings is proposed . unsup-SimCSE is biased because of the length information encoded into the sentence embeds .
Approach: They propose a new unsupervised sentence embedding method that uses dropout to obtain positive pairs from a pre-trained Transformer encoder.
Outcome: The proposed method outperforms the state-of-the-art unsup-SimCSE on a STS task.
RaP: Redundancy-aware Video-language Pre-training for Text-Video Retrieval (2022.findings-emnlp)

Copied to clipboard

Challenge: sparse sampling of videos suffers from inter-modal redundancy and visual redundancies . et al., 2021) proposes to sparsestly sample frames from videos to alleviate temporal redundance .
Approach: They propose to use sparse sampling to alleviate temporal redundancy in videos . they propose to penalize high-redundant video patches and text tokens .
Outcome: The proposed method improves on four benchmark datasets.
Resolving the Security-Auditability Dilemma with Auditable Latent Chain-of-Thought Alignment (2026.acl-long)

Copied to clipboard

Challenge: Extensive experiments show that ALCA reduces the success rate of adaptive jailbreak attacks by over 40% compared to strong baselines, while preserving performance.
Approach: They propose a framework that decouples internal reasoning from external output and allows the model to reconstruct its latent reasoning into human-readable text for supervision under specific guidance.
Outcome: The proposed framework reduces the success rate of adaptive jailbreak attacks by over 40% compared to baselines while preserving performance.
FAIRGAMER: Evaluating Social Biases in LLM-Based Video Game NPCs (2026.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) have enhanced or replaced traditional non-player characters in video games.
Approach: They propose a benchmark to evaluate social biases across three interaction patterns: transaction, cooperation, and competition.
Outcome: The proposed benchmark assesses four bias types across transaction, cooperation, and competition using a novel metric, FairMCV.
Breaking the Hourglass Phenomenon of Residual Quantization: Enhancing the Upper Bound of Generative Retrieval (2024.emnlp-industry)

Copied to clipboard

Challenge: Generative retrieval (GR) is a transformative paradigm in search and recommender systems . however, data sparsity and long-tailed distribution hinder the full utilization of GR .
Approach: They propose a method to reduce the "Hourglass" phenomenon in RQ-SID where codebook tokens become overly concentrated.
Outcome: The proposed methods improve retrieval efficiency and generalization capabilities.
FABLE: Fine-grained Fact Anchoring for Unstructured Model Editing (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for model editing memorize text holistically without reliable fine-grained fact access.
Approach: They propose a hierarchical framework that decouples fine-grained fact injection from holistic text generation.
Outcome: The proposed framework significantly improves fine-grained question answering while maintaining state-of-the-art holistic editing performance.
Smoothed Contrastive Learning for Unsupervised Sentence Embedding (2022.coling-1)

Copied to clipboard

Challenge: Unsupervised contrastive sentence embedding models use InfoNCE loss function . increasing batch size leads to performance degradation when it exceeds threshold .
Approach: They propose a simple smoothing strategy upon the InfoNCE loss function to reduce the number of false-negative pairs in a batch without increasing the batch size.
Outcome: The proposed smoothing strategy improves unsupervised SimCSE on semantic similarity tasks.
AdvancedIF: Rubric-Based Benchmarking and Reinforcement Learning for Advancing LLM Instruction Following (2026.acl-long)

Copied to clipboard

Challenge: Recent advances in large language models (LLMs) have shown impressive performance on a range of tasks, yet advanced instruction following (IF) remains a significant challenge.
Approach: They propose a benchmark that features over 1,600 prompts and expert-curated rubrics that assess LLMs’ ability to follow complex, multi-turn, and system-level instructions.
Outcome: The proposed framework improves instruction-following abilities of large language models, achieving a 6.7% gain on AdvancedIF and strong results on public benchmarks.
OpenRLHF: A Ray-based Easy-to-use, Scalable and High-performance RLHF Framework (2025.emnlp-demos)

Copied to clipboard

Challenge: Existing RLHF frameworks face inference bottlenecks and complexity barriers restricting their accessibility for newcomers.
Approach: They propose an open-source RLHF framework that can be used to train large language models.
Outcome: The proposed framework achieves superior training efficiency with speedups ranging from 1.22 to 1.68 across different model sizes compared to state-of-the-art frameworks, while requiring significantly fewer lines of code for implementation.
InfoCSE: Information-aggregated Contrastive Learning of Sentence Embeddings (2022.findings-emnlp)

Copied to clipboard

Challenge: Existing studies on contrastive learning for sentence embeddings are weak . researchers have started to use contrastive training to learn better unsupervised sentences.
Approach: They propose an information-aggregated contrastive learning framework for learning unsupervised sentence embeddings.
Outcome: The proposed framework outperforms SimCSE on several benchmark datasets w.r.t the semantic text similarity task.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations