Papers by Feng Xiong

22 papers
POSITION BIAS MITIGATES POSITION BIAS: Mitigate Position Bias Through Inter-Position Knowledge Distillation (2025.emnlp-main)

Copied to clipboard

Challenge: Positional bias (PB) manifests as non-uniform sensitivity across contextual locations . previous studies have addressed PB by modifying the underlying architectures or employing extensive contextual awareness training.
Approach: They propose a position-to-position knowledge distillation framework that leverages position-induced disparities to counteract PB.
Outcome: The proposed framework reduces positional bias and improves performance on retrieval and reasoning tasks.
Efficient Multi-Agent System Training with Data Influence-Oriented Tree Search (2026.acl-long)

Copied to clipboard

Challenge: Large Language Model (LLM) based multi-agent systems (MAS) have high potential for tackling complex tasks through collaborative intelligence.
Approach: They propose a framework that incorporates influence scores to guide tree search and data selection in data synthesis.
Outcome: The proposed framework incorporates influence scores to guide tree search and data selection in data synthesis.
Towards Human-aligned Evaluation for Linear Programming Word Problems (2024.lrec-main)

Copied to clipboard

Challenge: Existing evaluation methodologies for MWPs diverge from human judgment and face challenges in recognizing mathematically equivalent answers.
Approach: They propose an evaluation metric rooted in graph edit distance that features benefits such as permutation invariance and more accurate program equivalence identification.
Outcome: The proposed evaluation metric features benefits such as permutation invariance and more accurate program equivalence identification.
Learning Disentangled Semantic Representations for Zero-Shot Cross-Lingual Transfer in Multilingual Machine Reading Comprehension (2022.acl-long)

Copied to clipboard

Challenge: Existing methods to zero-shot transfer knowledge from rich-resource to low-resourced languages are limited due to linguistic discrepancies in different languages.
Approach: They propose a multilingual MRC framework equipped with a Siamese Semantic Disentanglement Model to disassociate semantics from syntax in models learned by multilingual pre-trained models.
Outcome: The proposed model disassociates semantics from syntax in multilingual models.
VPL: Visual Proxy Learning Framework for Zero-Shot Medical Image Diagnosis (2024.findings-emnlp)

Copied to clipboard

Challenge: Insufficient medical text precision and the modal disparity between text and vision spaces pose challenges for vision-language models like CLIP.
Approach: They propose a visual proxy learning framework that combines a text refinement module and a stable Sinkhorn algorithm to enhance the diagnostic performance.
Outcome: The proposed model outperforms the state-of-the-art CLIP inference by 1.69% to 15.31% on five datasets covering various diseases.
Visually-Guided Policy Optimization for Multimodal Reasoning (2026.acl-long)

Copied to clipboard

Challenge: Existing RLVRs lack visual faithfulness due to text-dominated reasoning . a novel framework to reinforce visual focus during policy optimization is proposed .
Approach: They propose a framework to reinforce visual focus during policy optimization using visual attention compensation mechanism.
Outcome: The proposed framework exhibits better visual activation and superior performance in multimodal reasoning and visual-dependent tasks.
SocialEval: Evaluating Social Intelligence of Large Language Models (2025.acl-long)

Copied to clipboard

Challenge: Existing work on LLMs does not address their social intelligence (SI) and their discrepancy with humans.
Approach: They propose a script-based bilingual SI benchmark that integrates outcome-oriented goal achievement evaluation and process-oriented interpersonal ability evaluation by manually crafting narrative scripts.
Outcome: The proposed model is based on a script-based bilingual evaluation paradigm that integrates outcome- and process-oriented evaluation by manually crafting narrative scripts.
PKAG-DDI: Pairwise Knowledge-Augmented Language Model for Drug-Drug Interaction Event Text Generation (2025.acl-long)

Copied to clipboard

Challenge: Drug-drug interactions arise when multiple drugs are administered concurrently.
Approach: They propose a pairwise knowledge-augmented generative method for DDIE text generation that integrates biological functions from a knowledge set into a language model.
Outcome: The proposed method outperforms existing methods in DDIE text generation on two professional datasets.
FOFO: A Benchmark to Evaluate LLMs’ Format-Following Capability (2024.acl-long)

Copied to clipboard

Challenge: Existing benchmarks fail to assess large language models’ format-following proficiency adequately.
Approach: They propose a benchmark to evaluate large language models' ability to follow complex, domain-specific formats.
Outcome: The proposed framework evaluates large language models' ability to follow complex, domain-specific formats across open-source and closed-source models.
AlignedCoT: Prompting Large Language Models via Native-Speaking Demonstrations (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing LLMs are delicate and elusive in prompt words and styles.
Approach: They propose an LLM-acquainted prompting technique that includes proficient "native-speaking" they propose to use in-context learning to prompt LLMs to perform high-performance reasoning .
Outcome: The proposed technique achieves step-wise prompts in zero-shot scenarios while maintaining the prompt quality.
HS-STaR: Hierarchical Sampling for Self-Taught Reasoners via Difficulty Estimation and Budget Reallocation (2025.emnlp-main)

Copied to clipboard

Challenge: Recent studies have incorporated reward models to guide response selection or decoding, aiming to obtain higher-quality data.
Approach: They propose a Hierarchical Sampling framework for self-taught reasoners that allocates a fixed sampling budget to problem boundary-level problems and then reallocates the remaining budget toward high-utility problems during a re-sampling phase.
Outcome: The proposed framework outperforms baseline models without additional sampling budgets across multiple reasoning benchmarks and backbone LLMs.
Improving Factual Consistency of News Summarization by Contrastive Preference Optimization (2024.findings-emnlp)

Copied to clipboard

Challenge: Recent advances in news summarization have created problems with “hallucinations” that are factually inconsistent with the source text.
Approach: They propose to disentangle LLMs’ propensities to generate faithful and fake content by adopting a probing-based specific training method to improve their capacity of distinguishing two types of propensity.
Outcome: The proposed method disentangles LLMs’ propensities to generate faithful and fake content and improves their ability to distinguish between two types of propensity.
Fusing Label Embedding into BERT: An Efficient Improvement for Text Classification (2021.findings-acl)

Copied to clipboard

Challenge: Existing methods to improve text classification performance of pre-trained models have been used to improve their performance.
Approach: They propose a method for improving BERT's performance by using a label embedding technique while keeping almost the same computational cost.
Outcome: The proposed method improves BERT's performance on six text classification benchmark datasets while keeping almost the same computational cost.
Re-embedding Difficult Samples via Mutual Information Constrained Semantically Oversampling for Imbalanced Text Classification (2021.emnlp-main)

Copied to clipboard

Challenge: Existing frameworks for imbalanced text classification can generate anchor instances for difficult samples . difficult samples are hard to classify as they are embedded into an overlapping semantic region with the majority class.
Approach: They propose a Mutual Information constrained Semantically Oversampling framework that generates anchor instances for difficult samples to help the backbone network determine the re-embedding position of a non-overlapping representation.
Outcome: The proposed framework can generate anchor instances to help classifiers achieve significant improvements over baselines on a variety of imbalanced text classification tasks.
Reasoning Over Space: Enabling Geographic Reasoning for LLM-Based Generative Next POI Recommendation (2026.acl-long)

Copied to clipboard

Challenge: Existing LLM-based recommenders lack explicit modeling of geographic signals . without explicit modeling geographic signals, recommenders struggle to capture core mobility patterns .
Approach: They propose a framework that utilizes geography as a decision variable within the reasoning process.
Outcome: The proposed framework achieves over 10% relative gains in hit rate over strongest LLM-based baselines and improves cross-city transfer.
Xinference: Making Large Model Serving Easy (2024.emnlp-demo)

Copied to clipboard

Challenge: Open-source large models are rapidly catching up with the closed-source models . however, many current inference tools are not as simple and convenient to use.
Approach: They develop an open-source library to simplify the deployment and management of large models.
Outcome: The proposed library outperforms open-source models and offers high throughput and low latency.
Unsupervised Out-of-Domain Detection via Pre-trained Transformers (2021.acl-long)

Copied to clipboard

Challenge: Prior work on out-of-domain detection requires in-domain task labels and is limited to supervised classification scenarios.
Approach: They propose a method to construct out-of-domain detectors efficiently using pre-trained transformers.
Outcome: The proposed method greatly improves out-of-domain detection ability in a more general scenario.
xLAM: A Family of Large Action Models to Empower AI Agent Systems (2025.naacl-long)

Copied to clipboard

Challenge: Autonomous agents powered by large language models (LLMs) have attracted significant research interest, but there are few standards for developing specialized models for agent tasks.
Approach: They propose a series of large action models with dense and mixture-of-expert architectures that unifies, augments, and synthesizes diverse datasets to enhance agent generalizability and performance.
Outcome: The proposed models outperform GPT-4, Claude-3, and many other models in terms of tool use and outperformed GPT-based models on multiple agent ability benchmarks.
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.
CARE-STaR: Constraint-aware Self-taught Reasoner (2025.findings-acl)

Copied to clipboard

Challenge: Recent research on instruction following has demonstrated that LLMs can handle complex instructions.
Approach: They propose to assign constraints to different levels of constraints in instructions . they use chain-of-thought and self-taught reasoner methods to identify constraints .
Outcome: The proposed method outperforms supervised fine-tuning (SFT) on three instruction-following benchmarks.
TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process Rewards (2026.acl-long)

Copied to clipboard

Challenge: Existing approaches to training multi-turn attackers to probe model safety vulnerabilities rely on turn-level optimization, which is insufficient for learning long-term attack strategies.
Approach: They propose a multi-turn reinforcement learning problem that optimizes the harmfulness of the final-turn response as the outcome reward.
Outcome: The proposed approach improves attack success rates across multiple models and benchmarks, highlighting the effectiveness of the proposed approach.

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