Papers by Xintong Li

24 papers
PaddleSpeech: An Easy-to-Use All-in-One Speech Toolkit (2022.naacl-demo)

Copied to clipboard

Challenge: PaddleSpeech is an open-source speech toolkit that supports speech-to-text and text-to speech tasks.
Approach: They describe the design philosophy and core architecture of PaddleSpeech to support several essential speech-to-text and text-to speech tasks.
Outcome: The proposed framework achieves competitive or state-of-the-art performance on various speech datasets and implements the most popular methods.
Explainable Chain-of-Thought Reasoning: An Empirical Analysis on State-Aware Reasoning Dynamics (2025.findings-emnlp)

Copied to clipboard

Challenge: Recent advances in chain-of-thought prompting have demonstrated the ability of large language models to perform multi-step reasoning.
Approach: They propose a framework to analyze latent dynamics of CoT trajectories for interpretability . they segment generated CoT into discrete reasoning steps and abstract each step into a spectral embedding based on token-level Gram matrices .
Outcome: The proposed framework segments generated CoT steps into discrete reasoning steps, abstracts each step into a spectral embedding based on token-level Gram matrices, and clusters these embeddements into semantically meaningful latent states.
Mitigating Visual Knowledge Forgetting in MLLM Instruction-tuning via Modality-decoupled Gradient Descent (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing fine-tuning and continual learning methods compress visual representations and emphasize task alignment over visual retention.
Approach: They propose a modality-decoupled gradient descent (MDGD) that regulates gradient updates to preserve effective rank of visual features and explicitly disentangles visual learning from task-specific alignment.
Outcome: The proposed model reduces visual forgetting and improves visual retention . it disentangles visual learning from task-specific alignment and preserves effective rank .
M2PA: A Multi-Memory Planning Agent for Open Worlds Inspired by Cognitive Theory (2025.findings-acl)

Copied to clipboard

Challenge: Open-world planning poses a challenge due to complex environments and task diversity . recent work shows that large language models (LLMs) lack the ability to connect to agents' experiences .
Approach: They propose an open-world multi-memory planning agent that combines large language models with human-like multi-mesh systems to leverage their strengths.
Outcome: The proposed agent outperforms state-of-the-art agents on 50 Minecraft tasks in zero-shot learning.
Toward Multi-Session Personalized Conversation: A Large-Scale Dataset and Hierarchical Tree Framework for Implicit Reasoning (2025.emnlp-main)

Copied to clipboard

Challenge: Existing long-term open-domain dialogue datasets lack complex, real-world personalization and fail to capture implicit reasoning.
Approach: They propose a large-scale long-term dataset with 2,500 examples containing approximately 100 conversation sessions to study implicit reasoning in personalized dialogues.
Outcome: The proposed model improves the ability of LLMs to reason over long-term conversations with implicit contextual dependencies.
From Selection to Generation: A Survey of LLM-based Active Learning (2025.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) have been used for selection and training of data for active learning.
Approach: They propose an intuitive taxonomy that categorizes LLM-based active learning techniques and discuss the transformative roles they can play in the active learning loop.
Outcome: The proposed model can generate entirely new data instances and provide more cost-effective annotations with fewer labeled data instances.
Bit-by-Bit: Progressive QAT Strategy with Outlier Channel Splitting for Stable Low-Bit LLMs (2026.acl-long)

Copied to clipboard

Challenge: Existing approaches to training LLMs at ultra-low precisions suffer from convergence instability and substantial training costs.
Approach: They propose a progressive QAT framework with outlier channel splitting to address these issues . they use nested structure of integer quantization grids to enable a "train once, deploy any precision" paradigm .
Outcome: The proposed framework outperforms baselines on both Llama2/3 and W2A16, with an 11 speedup over BF16.
EverMemOS: A Self-Organizing Memory Operating System for Structured Long-Horizon Reasoning (2026.acl-long)

Copied to clipboard

Challenge: Existing memory systems for LLMs store isolated records and retrieve fragments . Existing systems store isolated data and fragments, limiting their ability to consolidate evolving experience and resolve conflicts.
Approach: They propose an engram-inspired memory operating system that implements an 'engram'-inspired lifecycle for computational memory.
Outcome: Experiments on LoCoMo, LongMemEval, and PersonaMeM-v2 show that EverMemeOS outperforms state-of-the-art methods on memory-augmented reasoning tasks.
PiKGL: Leveraging Pruned Knowledge Graphs for Explainable Stance Detection (2026.tacl-1)

Copied to clipboard

Challenge: Experimental results demonstrate that a Pruned interpretable knowledge Graph Learning framework for explainable stance detection is state-of-the-art for social media stance prediction.
Approach: They propose a Pruned interpretable knowledge Graph Learning framework for explainable stance detection that incorporates commonsense knowledge and prunes redundant information to ensure precision and minimize noise.
Outcome: The proposed framework achieves state-of-the-art on three public datasets.
On the Word Alignment from Neural Machine Translation (P19-1)

Copied to clipboard

Challenge: Prior researches suggest that neural machine translation (NMT) captures word alignment through its attention mechanism, however, attention may fail to capture word alignment for some NMT models.
Approach: They propose two methods to induce word alignment which are general and agnostic to specific NMT models.
Outcome: The proposed methods induce much better word alignment than attention.
Understanding and Improving Hidden Representations for Neural Machine Translation (N19-1)

Copied to clipboard

Challenge: Existing studies have explored some methods for understanding hidden representations, but they have not sought to improve the translation quality rationally according to their understanding.
Approach: They propose to construct a sequence of nested relative tasks and measure the feature generalization ability of the learned hidden representation over these tasks.
Outcome: The proposed methods achieve consistent improvements (up to +1.3 BLEU) on two widely-used datasets.
Regularized Context Gates on Transformer for Machine Translation (2020.acl-main)

Copied to clipboard

Challenge: Context gates are effective to control the contributions from the source and target contexts in the recurrent neural network (RNN) based neural machine translation.
Approach: They propose a method to identify source and target contexts and introduce a gate mechanism to control the contributions from source and targets in the advanced Transformer architecture.
Outcome: The proposed model achieves an averaged gain of 1.0 BLEU score over a strong transformer baseline.
LoRATK: LoRA Once, Backdoor Everywhere in the Share-and-Play Ecosystem (2025.findings-emnlp)

Copied to clipboard

Challenge: distributing LLMs without a proven track record like ‘meta-llama‘ or ‘qwen‘ rarely gains community traction.
Approach: They propose a simple, efficient, yet specific recipe for a backdoor LoRA to be injected into task-enhancing LoRAs and examine the mechanisms of such infections.
Outcome: The proposed model allows attackers to scale the distribution of compromised LoRAs with minimal effort by leveraging the rich pool of shared LoRA assets.
Open-world Multi-label Text Classification with Extremely Weak Supervision (2024.emnlp-main)

Copied to clipboard

Challenge: Similar single-label XWS settings cannot be easily adapted for multi-l label classification.
Approach: They propose a novel method for open-world multi-label text classification under extremely weak supervision where the user provides a brief description without any labels or ground-truth label space.
Outcome: The proposed method exhibits a remarkable increase in ground-truth label space coverage on various datasets.
QaRL: Rollout-Aligned Quantization-Aware RL for Fast and Stable Training under Training–Inference Mismatch (2026.findings-acl)

Copied to clipboard

Challenge: Recent work has shown that reinforcement learning with simple rule-based reward functions (RLVR) can induce emergent reasoning behaviors and yield gains in challenging domains such as math problem solving.
Approach: They propose a rollout-alignment-quantization-aware RL which aligns training-side forward with the quantized rollout to minimize mismatch.
Outcome: The proposed approach outperforms quantized-rollout training by +5.5 on Qwen3-30B-A3B MoE for math problems while maintaining low-bit throughput.
CoMMIT: Coordinated Multimodal Instruction Tuning (2025.emnlp-main)

Copied to clipboard

Challenge: et al., 2024) show that multimodal instruction tuning is more effective than baselines.
Approach: They propose a multimodal balance coefficient that enables quantitative measurement of the balance of learning . they propose auxiliary regularization on the gradient to promote updating with larger step sizes .
Outcome: The proposed method is more effective than baselines in MLLM instruction tuning.
MelTrim: Coarse-to-Fine Data Pruning for Speech Classification (2026.findings-acl)

Copied to clipboard

Challenge: Unlike image or text classification, speech classification tasks are particularly challenging due to the difficulty in capturing the acoustic, semantic, and contextual representations.
Approach: They propose a dataset pruning method that coarsely filters redundant samples using DBSCAN clustering on Mel-Frequency Cepstral Coefficients (MFCC) features.
Outcome: The proposed method achieves 49.5% improvement in WA on the MEAD dataset and 41.9% reduction in EER on speaker identification tasks.
GUI Agents: A Survey (2025.findings-acl)

Copied to clipboard

Challenge: Large Foundation Models (LFMs) have transformed the landscape of AI research and day-to-day life.
Approach: They propose a framework that delineates GUI agents' perception, reasoning, planning, and acting capabilities.
Outcome: The proposed framework delineates their perception, reasoning, planning, and acting capabilities.
HAT: Hallucination Annotation for Translation (2026.acl-long)

Copied to clipboard

Challenge: Hallucinations in machine translation (MT) outputs are prone to hallucination, authors say . lack of high-quality benchmarks for halluciation detection has hindered MT deployments .
Approach: They propose a dataset that provides annotated hallucination distributions and benchmarks . they use 350,959 span-level annotations across 38 language pairs to analyze hallucis a MT output .
Outcome: The proposed dataset provides high-quality benchmarks for hallucination detection in machine translation . the dataset includes 350,959 span-level annotated samples across 38 language pairs .
Building Adaptive Acceptability Classifiers for Neural NLG (2021.emnlp-main)

Copied to clipboard

Challenge: Existing approaches to generate synthetic data using simple sentence transformations and/or model-based techniques may not generate realistic error samples with respect to the NLG models.
Approach: They propose a framework to train models to classify acceptability of responses generated by natural language generation models using a 2-stage approach . they use existing sentence transformations to generate samples that better resemble the output of the generation models.
Outcome: The proposed approach outperforms existing techniques and can be used in few-shot settings using self-training.
Metagent-P: A Neuro-Symbolic Planning Agent with Metacognition for Open Worlds (2025.findings-acl)

Copied to clipboard

Challenge: Recent advances in large language models (LLMs) show promising potential through their world knowledge and language processing capabilities in open-world planning.
Approach: They propose a framework that integrates the world knowledge of large language models, symbolic reasoning capabilities of cognitive architectures, and metacognition to improve experience utilization.
Outcome: The proposed framework outperforms current state-of-the-art methods in Minecraft and reduces the average replanning counts by 34% and exceeds the human success rate by 18.96%.
SceneAlign: Aligning Multimodal Reasoning to Scene Graphs in Complex Visual Scenes (2026.acl-long)

Copied to clipboard

Challenge: Existing preference-based approaches fail to address this challenge by exploiting language priors to bypass visual grounding.
Approach: They propose a framework that leverages scene graphs as structured visual information to perform controllable structural interventions.
Outcome: The proposed framework improves answer accuracy and reasoning faithfulness across seven visual reasoning benchmarks.
CogSteer: Cognition-Inspired Selective Layer Intervention for Efficiently Steering Large Language Models (2025.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) achieve excellent performance through pretraining on extensive data.
Approach: They propose an efficient selective layer intervention based on parameter-efficient fine-tuning methods to select the optimal steering layer to modulate LLM semantics.
Outcome: The proposed approach is based on a model-agnostic framework and is safe to deploy.
Target Foresight Based Attention for Neural Machine Translation (N18-1)

Copied to clipboard

Challenge: Empirical experiments on Chinese-to-English and Japanese-to English datasets show that the proposed attention model delivers significant improvements in terms of alignment error rate and BLEU.
Approach: They propose to explicitly access the target foresight word in the attention model to improve alignment and translation accuracy.
Outcome: Empirical results show that the proposed model improves alignment error rate and BLEU on Chinese-to-English and Japanese-toEnglish datasets.

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