Papers by Yifei Xu
Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search (2026.findings-acl)
Copied to clipboard
Yifei Zhang, Xu Yang, Xiao Yang, Bowen Xian, Qizheng Li, Shikai Fang, Jingyuan Li, Jian Wang, Minrui Xu, Yuge Zhang, Weiqing Liu, Jiang Bian
| Challenge: | LLM-based agents for machine learning engineering rely on tree search to rank candidates. |
| Approach: | They propose an LLM-based agent that operationalizes gradient-based optimization. |
| Outcome: | The proposed agent achieves a state-of-the-art 35.1% any-medal rate on MLE-Bench with a limited budget on a single GPU. |
SLOT: Structuring the Output of Large Language Models (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Structured outputs are essential for large language models (LLMs) but often deviate from predefined schemas hampering reliable application development. |
| Approach: | They propose a model-agnostic approach that transforms unstructured LLM outputs into precise structured formats. |
| Outcome: | The proposed model-agnostic approach transforms unstructured LLM outputs into precise structured formats. |
Chain of Ideas: Revolutionizing Research Via Novel Idea Development with LLM Agents (2025.findings-emnlp)
Copied to clipboard
Long Li, Weiwen Xu, Jiayan Guo, Ruochen Zhao, Xingxuan Li, Yuqian Yuan, Boqiang Zhang, Yuming Jiang, Yifei Xin, Ronghao Dang, Yu Rong, Deli Zhao, Tian Feng, Lidong Bing
| Challenge: | Existing methods for idea generation either trivially prompt LLMs or expose LLM to extensive literature without indicating useful information. |
| Approach: | They propose a chain-of-ideas agent that organizes literature in a chains structure . they propose evaluating idea-generation methods from different perspectives . |
| Outcome: | The proposed agent outperforms existing methods and matches human quality in idea generation. |
Locomo-Plus: Beyond-Factual Cognitive Memory Evaluation Framework for LLM Agents (2026.acl-long)
Copied to clipboard
| Challenge: | Existing benchmarks and evaluation protocols focus on surface-level factual recall. |
| Approach: | They propose a benchmark for assessing cognitive memory under cue–trigger semantic disconnect. |
| Outcome: | The proposed framework reveals failures not captured by existing benchmarks. |
IPR: Intelligent Prompt Routing with User-Controlled Quality-Cost Trade-offs (2025.emnlp-industry)
Copied to clipboard
Aosong Feng, Balasubramaniam Srinivasan, Yun Zhou, Zhichao Xu, Kang Zhou, Sheng Guan, Yueyan Chen, Xian Wu, Ninad Kulkarni, Yi Zhang, Zhengyuan Shen, Dmitriy Bespalov, Soumya Smruti Mishra, Yifei Teng, Darren Yow-Bang Wang, Haibo Ding, Lin Lee Cheong
| Challenge: | Existing systems require users to manually select models or employ rigid routing rules that fail to capture the continuous spectrum of query complexity. |
| Approach: | They propose a quality-constrained intelligent prompt routing framework that automatically selects optimal models based on predicted response quality and user-specified tolerance levels. |
| Outcome: | The proposed framework achieves 43.9% cost reduction while maintaining quality parity with strongest model in the Claude family and processes requests with sub-150ms latency. |
MoLAN: A Unified Modality-Aware Noise Dynamic Editing Framework for Multimodal Sentiment Analysis (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to multimodal sentiment analysis treat entire modality as an independent unit for feature enhancement or denoising, which often suppresses redundant noise at the cost of weakening critical information. |
| Approach: | They propose a ModaLity-aware noise dynAmic editiNg framework that performs modality-awful block partitioning by dividing features of each modality into multiple blocks. |
| Outcome: | Experiments on five models and four datasets show that MoLAN+ achieves the state-of-the-art performance. |
MUR: Momentum Uncertainty guided Reasoning for Large Language Models (2026.acl-long)
Copied to clipboard
Hang Yan, Fangzhi Xu, Rongman Xu, Yifei Li, Jian Zhang, Haoran Luo, Xiaobao Wu, Anh Tuan Luu, Haiteng Zhao, Qika Lin, Jun Liu
| Challenge: | Existing methods for optimizing reasoning quality are limited by overthinking. |
| Approach: | They propose a method that allocates thinking budgets to critical reasoning steps by tracking and aggregating step-wise uncertainty over time. |
| Outcome: | The proposed method reduces computation by over 45% on average while improving accuracy by 0.33–3.46%. |
FIRST: Faster Improved Listwise Reranking with Single Token Decoding (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing listwise LLMs lack efficiency as they provide ranking output in the form of a generated ordered sequence of candidate passage identifiers. |
| Approach: | They propose a listwise LLM reranking approach that leverages the first generated identifier to obtain a ranked ordering of the candidates. |
| Outcome: | The proposed approach accelerates inference by 50% while maintaining robust ranking performance with gains across BEIR benchmark. |
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. |
KC-ISA: An Implicit Sentiment Analysis Model Combining Knowledge Enhancement and Context Features (2022.coling-1)
Copied to clipboard
| Challenge: | Existing research results on explicit sentiment analysis are limited . implicit sentiment analysis is a process of analyzing text based on whether it contains explicit sentiment words. |
| Approach: | They propose a model that integrates external knowledge and contextual features . they use a knowledge graph to supplement implicit sentiment expression . |
| Outcome: | The proposed model can achieve better results on the SMP2019 implicit sentiment analysis dataset. |
CtrlNews: LLM-based Multi-Agent Controllable News Writing via Knowledge Gravitational Field (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Current approaches to news writing rely on superficially retrieved information and oversimplified knowledge enumeration resulting in shallow, repetitive, and unordered outputs. |
| Approach: | They propose an LLM-based multi-agent controllable news writing framework called CtrlNews . they propose a fine-grained viewpoint control mechanism to regulate bias, emotion, and exaggeration attributes. |
| Outcome: | The proposed framework simulates expert questioning through automated role assignment and question generation followed by a three-layer hierarchical gravitational graph iteratively refined via expansion-reflection cycles. |
Adaptive Reinforcement Tuning Language Models as Hard Data Generators for Sentence Representation (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods use contrastive learning (CL) to learn effective sentence representations, but require extensive human annotation. |
| Approach: | They propose a reinforcement learning approach for fine-tuning small-parameter LLMs to generate high-quality hard contrastive data without human feedback. |
| Outcome: | The proposed method achieves state-of-the-art on seven semantic text similarity tasks. |
CogTaskonomy: Cognitively Inspired Task Taxonomy Is Beneficial to Transfer Learning in NLP (2022.acl-long)
Copied to clipboard
| Challenge: | Existing frameworks for transfer learning across tasks in natural language processing are CRA and CNM . CRA uses a correlation between brain-activity measurement and computational modeling to estimate task similarity with sentence representations. |
| Approach: | They propose a cognitively inspired framework to learn taxonomy for NLP tasks . they use Cognitive Representation Analytics and Cognitive-Neural Mapping . |
| Outcome: | The proposed framework can guide transfer learning across tasks in natural language processing without exhaustive pairwise task transferring. |
Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations (2024.acl-long)
Copied to clipboard
| Challenge: | Existing methods for process-oriented math reward models rely on manual annotation. |
| Approach: | They propose a process-oriented math process reward model called Math-shepherd which assigns a reward score to each step of math problem solutions. |
| Outcome: | The proposed model breaks the bottleneck of manual supervision in two scenarios. |
DeepSpecs: Expert-Level Question Answering in 5G (2026.findings-acl)
Copied to clipboard
| Challenge: | 3GPP standards define the technical design and implementation of 5G systems . expert-level questions require navigating thousands of pages of cross-referenced standards . |
| Approach: | They propose a standard-native retrieval-augmented generation system that can answer 5G questions . they use SpecDB, ChangeDB, TDocDB and a metadata-rich retrieval system to do this . |
| Outcome: | The proposed solution outperforms base models and state-of-the-art RAG systems in QA datasets . expert-level queries require navigating thousands of pages of cross-referenced standards . |
Hard2Verify: A Step-Level Verification Benchmark for Open-Ended Frontier Math (2026.acl-long)
Copied to clipboard
| Challenge: | Large language model (LLM)-based reasoning systems have recently achieved gold medal-level performance in the IMO 2025 competition . |
| Approach: | They propose a human-annotated step-level verification benchmark that measures step- level verifiers at the frontier. |
| Outcome: | The proposed benchmark outperforms closed-source models in step-level verification and the impact of scaling verifier compute. |
Does Context Matter? ContextualJudgeBench for Evaluating LLM-based Judges in Contextual Settings (2025.acl-long)
Copied to clipboard
| Challenge: | Contextual evaluation is challenging for state-of-the-art judge models . evaluation criteria are often conditional and dependent on practitioner priorities . |
| Approach: | They propose a judge benchmark that evaluates large language models as judges in contexts . they use human annotations and model-based perturbations to build the benchmark . |
| Outcome: | The proposed benchmark aims to evaluate large language models in contexts with 2,000 challenging response pairs. |
FedLFC: Towards Efficient Federated Multilingual Modeling with LoRA-based Language Family Clustering (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing frameworks for multilingual modeling face communication costs and parameter interference conflicts. |
| Approach: | They propose a communication-efficient federated learning framework with low-rank adaptation and language family clustering for Multilingual Modeling (MM) they maintain the weights of the base model, updating the lightweight Low-rank adapt parameters to minimize communication costs. |
| Outcome: | The proposed model outperforms the baseline models in performance and reduces communication overhead. |
RATION: Entropy-Driven Task-Adaptive Visual Attention Allocation Framework for Multimodal Reasoning (2026.findings-acl)
Copied to clipboard
| Challenge: | Prior studies have focused on strengthening multimodal reasoning by improving representation alignment or increasing computation, but these methods do not characterize the differences in visual demands across tasks. |
| Approach: | They propose an entropy-driven task-adaptive visual attention allocation framework that uses visual attention entropic as a control signal to dynamically allocate attention according to task demands. |
| Outcome: | The proposed framework achieves consistent performance gains across diverse reasoning tasks, datasets, and models, providing a clear direction toward more reliable multimodal reasoning. |