Papers by Jindong Gu

14 papers
Magnet: Multi-turn Tool-use Data Synthesis and Distillation via Graph Translation (2025.acl-long)

Copied to clipboard

Challenge: Large language models have been shown to be effective in multi-turn interactions . however, their performance may be limited in complex, multi-turned interactions involving users and multiple tools.
Approach: They propose a framework for synthesizing high-quality training trajectories to enhance the function calling capability of large language model agents in multi-turn conversations with humans.
Outcome: The proposed model outperforms the teacher model by 68.01 on BFCL-v3 and 73.30 on ToolQuery.
LLM Jailbreak Detection for (Almost) Free! (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for detecting jailbreak prompts entail significant computational costs .
Approach: They propose a free jailbreak detection method which scales logits by temperature to detect jailbreak prompts .
Outcome: The proposed method detects jailbreak prompts with no additional computational costs.
FocalPO: Enhancing Preference Optimizing by Focusing on Correct Preference Rankings (2025.acl-short)

Copied to clipboard

Challenge: Efficient preference optimization algorithms such as Direct Preference Optimization (DPO) have become a popular approach in aligning large language models with human preferences.
Approach: They propose a preference optimization variant that instead down-weighs misranked preference pairs and prioritizes enhancing the model’s understanding of pairs that it can already rank correctly.
Outcome: The proposed model outperforms DPO on benchmarks like Alpaca Eval 2.0 and Arena-Hard using mistral-base-7B and Llama-3-Instruct-8B with the introduced hyperparameter fixed.
Benchmarking Open-ended Audio Dialogue Understanding for Large Audio-Language Models (2025.acl-long)

Copied to clipboard

Challenge: Large Audio-Language Models (LALMs) have recently unlocked audio dialogue capabilities, enabling direct spoken exchanges with humans.
Approach: They propose to evaluate LALMs' open-ended audio dialogue ability in 3 general scenarios, 12 skills, 9 multilingual languages, and 4 categories of ambiguity handling.
Outcome: The proposed benchmark assesses the open-ended audio dialogue ability for LALMs in 3 general scenarios, 12 skills, 9 multilingual languages, and 4 categories of ambiguity handling.
PlanGEN: A Multi-Agent Framework for Generating Planning and Reasoning Trajectories for Complex Problem Solving (2025.emnlp-main)

Copied to clipboard

Challenge: Existing methods for natural planning lack constraint-guided iterative verification and adaptive selection . a recent study found that LLMs are not good at such planning.
Approach: They propose a model-agnostic and easily scalable agent framework with three key components: constraint, verification, and selection agents.
Outcome: The proposed framework improves inference-time algorithms on NATURAL PLAN and OlympiadBench benchmarks.
Flaw or Artifact? Rethinking Prompt Sensitivity in Evaluating LLMs (2025.emnlp-main)

Copied to clipboard

Challenge: a high prompt sensitivity has been widely accepted as a core limitation of large language models . a recent study suggests that prompt senescence may be an artifact of evaluation processes .
Approach: They examine whether prompt sensitivity is an inherent weakness or an artifact of evaluation . they find that heuristic evaluation methods overlook semantically correct responses . large language models have achieved remarkable success across a wide range of tasks .
Outcome: The proposed model is more robust to prompt templates than previously thought . the authors show that prompt sensitivity may be an artifact of evaluation rather than a flaw .
Reimagining Safety Alignment with An Image (2025.emnlp-main)

Copied to clipboard

Challenge: Existing approaches to large language models face inefficiency, fragility, or architectural constraints, resulting in inefficient performance and heightened over-refusal in cross-modal tasks.
Approach: They propose an optimization-driven visual prompt framework that enhances security and reduces over-refusal at the same time.
Outcome: The proposed framework enhances security and reduces over-refusal while maintaining robust safety while reducing unnecessary denials.
Multimodal Pragmatic Jailbreak on Text-to-image Models (2025.acl-long)

Copied to clipboard

Challenge: Existing jailbreaks for diffusion-based text-to-image models generate unsafe content . experimental results show that all tested models suffer from unsafe generation .
Approach: They propose a jailbreak that triggers diffusion-based text-to-image models to generate the image with visual text, resulting in unsafe content.
Outcome: The proposed model generates image with visual text, but the model is unsafe under such jailbreak.
Visual Question Decomposition on Multimodal Large Language Models (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for question decomposition focus on unimodal language models, but question decomposing capability of Multimodal Large Language Models (MLLMs) has yet to be explored.
Approach: They propose a finetuning dataset and a training objective for selective decomposition to enhance the model's question decomposing capability.
Outcome: The proposed dataset shows that existing models struggle to produce high-quality sub-questions.
Beyond Cross-Modal Alignment: Measuring and Leveraging Modality Gap in Vision-Language Models (2026.findings-acl)

Copied to clipboard

Challenge: a recent study shows that vision-language models have modality gaps that persist even in well-aligned models.
Approach: They propose a modality-dominance score to measure and leverage modality gaps . they propose automatic interpretability metrics to evaluate these features in a scalable manner .
Outcome: The proposed framework allows for training-free probing and editing methods for understanding model perception across genders and generating adversarial examples.
You Only Need One Single Token to Refine Safety Alignment (2026.findings-acl)

Copied to clipboard

Challenge: Excessive safety can lead to over-refusal, where models reject harmful-looking yet benign queries, severely limiting utility.
Approach: They propose a lightweight training-based approach that reshapes the distributions of harmful and benign samples within the model’s decision space by using a single-token prefix.
Outcome: The proposed approach can distinguish between harmful and benign samples while keeping the model frozen.
ECOLA: Enhancing Temporal Knowledge Embeddings with Contextualized Language Representations (2023.findings-acl)

Copied to clipboard

Challenge: Existing enhancement approaches cannot be applied to temporal knowledge graphs (tKGs) existing enhancement approaches assume knowledge embedding is time-independent, whereas entity embedded in tKG models evolves .
Approach: They propose to use textual data to enhance temporal knowledge embedding by Enhanced Temporal Knowledge Embeddings with Contextualized Language Representations (ECOLA) to evaluate ECOLA, they introduce three new datasets for training and evaluation.
Outcome: The proposed model significantly improves Hits@1 on the link prediction task.
Knowledge Control for Responsible Generative AI: Bridging Academia, Industry, and Society (2026.acl-tutorials)

Copied to clipboard

Challenge: This tutorial introduces the foundations of post-training knowledge control and showcases recent frontier methods.
Approach: This tutorial introduces the foundations of post-training knowledge control and showcases recent frontier methods.
Outcome: This tutorial introduces the foundations of post-training knowledge control and showcases recent frontier methods . key motivations and failure modes, harmful generation and stereotype reinforcement, are addressed . core methods such as machine unlearning, knowledge editing, and inference-time interventions are also included .
Can an Individual Manipulate the Collective Decisions of Multi-Agents? (2025.emnlp-main)

Copied to clipboard

Challenge: Recent studies show that coordinated multi-agent systems exhibit enhanced decision-making and reasoning abilities through collaboration.
Approach: They propose a framework that simulates agent interactions within a multi-agent system to generate adversarial samples and use them to manipulate the target agent in the target system.
Outcome: The proposed framework generates adversarial samples that are used to manipulate the target agent in the target system, misleading the system’s decision-making process.

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