Papers with automated framework
Divide-Conquer-Reasoning for Consistency Evaluation and Automatic Improvement of Large Language Models (2024.emnlp-industry)
Copied to clipboard
| Challenge: | Existing methods for evaluating the quality and consistency of text generated by Large Language Models are not effective. |
| Approach: | They propose a divide-conquer-reasoning approach to evaluate LLM-generated texts using a split-and-conquers evaluator and an automatic metric converter to facilitate this approach. |
| Outcome: | The proposed framework outperforms state-of-the-art methods by a large margin on multiple benchmarks and reduces 90% of output inconsistencies in one iteration. |
Benchmark Transparency: Measuring the Impact of Data on Evaluation (2024.naacl-long)
Copied to clipboard
| Challenge: | In this paper, we quantify the impact that data distribution has on the performance and evaluation of NLP models. |
| Approach: | They propose to use disproportional stratified sampling to measure the data distribution across 6 different dimensions to quantify model performance. |
| Outcome: | The proposed framework measures the data distribution across 6 different dimensions and shows that it is statistically significant and predicts model performance. |
VN-MTEB: Vietnamese Massive Text Embedding Benchmark (2026.findings-eacl)
Copied to clipboard
| Challenge: | a lack of large-scale test datasets makes it difficult to evaluate AI models before deploying them in real-world projects. |
| Approach: | They propose a Vietnamese benchmark for embedding models that leverages large language models and embeddable models to translate and filter samples from the Massive Multilingual Text Embedding Benchmark. |
| Outcome: | The proposed benchmark outperforms existing models in Vietnamese and English tasks with 41 datasets. |
Self-prompted Chain-of-Thought on Large Language Models for Open-domain Multi-hop Reasoning (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing open-domain question-answering methods lack quality assurance . existing methods lack scalability and poor diversity, hindering LLMs' capabilities . |
| Approach: | They propose an open-domain multi-hop reasoning framework to answer multi-choice questions . they propose an adaptive sampler for in-context selection and self-prompted inference . |
| Outcome: | The proposed framework surpasses the existing SOTA methods on large-scale datasets and doubles the zero-shot performance of small-scale LLMs. |
Understanding Conflicts in Multi-Objective Alignment through Reward Consistency (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing training pipelines still face alignment conflicts where optimizing for one objective degrades performance on others. |
| Approach: | They propose a reward-based criterion that approximates alignment conflicts via reward models. |
| Outcome: | The proposed framework improves harmlessness and helpfulness scores by 23.07% over the vanilla dataset. |
AutoTaskEval: Towards Domain-Specific and Fine-Grained Evaluation for LLMs (2026.acl-long)
Copied to clipboard
Qingqing Lyu, Linjuan Wu, Yongliang Shen, Hengwei Liu, Hao Li, Shengpei Jiang, Yin Zhang, Weiming Lu
| Challenge: | Existing automated approaches operate within fixed task schemas and often fail to autonomously discover new evaluation dimensions. |
| Approach: | They propose an automated framework that constructs domain-specific benchmarks directly from unstructured corpora using Bloom’s Taxonomy. |
| Outcome: | The proposed framework uncovers a broader and more fine-grained task space than expert-curated benchmarks while producing high-quality instances that preserve established model-level evaluation trends. |
Evian: Towards Explainable Visual Instruction-tuning Data Auditing (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing data filtering methods rely on coarse-grained scores that lack granularity to identify nuanced semantic flaws. |
| Approach: | They propose a "Decomposition-then-Evaluation" paradigm that breaks model responses into constituent cognitive components. |
| Outcome: | The proposed model outperforms models trained on larger datasets in three key areas . the authors show that Logical Coherence is the most critical factor in data quality evaluation . |
What Makes a Good Reasoning Chain? Uncovering Structural Patterns in Long Chain-of-Thought Reasoning (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in reasoning with large language models have popularized Long Chain-of-Thought (LCoT) a framework that converts sequential LCoTs into hierarchical tree structures enables deeper structural analysis of LLM reasoning. |
| Approach: | They propose a framework that converts sequential LCoTs into hierarchical tree structures and enables deeper structural analysis of LLM reasoning. |
| Outcome: | The proposed framework can be used to analyze LLM reasoning in a variety of tasks and models. |
XAutoLM: Efficient Fine-Tuning of Language Models via Meta-Learning and AutoML (2025.emnlp-main)
Copied to clipboard
Ernesto Luis Estevanell Valladares, Suilan Estevez-Velarde, Yoan Gutierrez, Andrés Montoyo, Ruslan Mitkov
| Challenge: | XAutoLM is a meta-learning-augmented framework that can be used to optimize discriminative and generative LM fine-tuning pipelines. |
| Approach: | They propose a meta-learning-augmented AutoML framework that reuses past experiences to optimize discriminative and generative LM fine-tuning pipelines efficiently. |
| Outcome: | XAutoLM surpasses zero-shot optimizer’s peak F1 on five of six tasks, reduces mean evaluation time of pipelines by up to 4.5x, and uncovers 50% more pipelines above zero- shot Pareto front. |
EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are expected to be trained to act as agents in various real-world environments, but access to real systems is often restricted and manually built sandboxes are hard to scale. |
| Approach: | They propose an automated framework for scalable tool-interaction environments via programmatic synthesis that synthesizes 191 environments and about 7K scenarios and applies them to Supervised Fine-Tuning and Reinforcement Learning for Qwen3 series models. |
| Outcome: | The proposed framework significantly improves LLMs’ ability to solve tasks in complex environments involving multi-turn, multi-tool interactions. |
Automatic Extraction of Rules Governing Morphological Agreement (2020.emnlp-main)
Copied to clipboard
Aditi Chaudhary, Antonios Anastasopoulos, Adithya Pratapa, David R. Mortensen, Zaid Sheikh, Yulia Tsvetkov, Graham Neubig
| Challenge: | Creating a descriptive grammar is an indispensable step for language documentation but it is tedious and time-consuming. |
| Approach: | They propose a framework for extracting a first-pass grammatical specification from raw text in a concise, human- and machine-readable format. |
| Outcome: | The proposed framework extracts a grammatical specification that is nearly equivalent to those created with large amounts of gold-standard annotated data. |
M-LongDoc: A Benchmark For Multimodal Super-Long Document Understanding And A Retrieval-Aware Tuning Framework (2025.emnlp-main)
Copied to clipboard
Yew Ken Chia, Liying Cheng, Hou Pong Chan, Maojia Song, Chaoqun Liu, Mahani Aljunied, Soujanya Poria, Lidong Bing
| Challenge: | Existing benchmarks for large multimodal models focus on short documents with less than 50 pages and are limited to extraction-based questions. |
| Approach: | They propose a retrieval-aware tuning approach to improve the accuracy of multimodal document reading by 4.6%. |
| Outcome: | The proposed framework improves the accuracy of model responses by 4.6% compared to existing benchmarks on documents with hundreds of pages and longer documents with more complex content. |
HoLLMwood: Unleashing the Creativity of Large Language Models in Screenwriting via Role Playing (2024.findings-emnlp)
Copied to clipboard
Jing Chen, Xinyu Zhu, Cheng Yang, Chufan Shi, Yadong Xi, Yuxiang Zhang, Junjie Wang, Jiashu Pu, Tian Feng, Yujiu Yang, Rongsheng Zhang
| Challenge: | Generative AI has demonstrated unprecedented creativity in the field of computer vision, yet such phenomena have not been observed in the realm of literary creation. |
| Approach: | They propose a framework for unleashing the creativity of large language models (LLMs) they assign LLMs to different roles involved in real-world scenario, they write . |
| Outcome: | The proposed framework outperforms baselines in terms of coherence, relevance, interestingness and overall quality on automatically generated screenplays. |
ResearchBench: Benchmarking LLMs in Scientific Discovery via Inspiration-Based Task Decomposition (2026.findings-acl)
Copied to clipboard
Yujie Liu, Zonglin Yang, Tong Xie, Jinjie Ni, Ben Gao, Yuqiang Li, Shixiang Tang, Wanli Ouyang, Erik Cambria, Dongzhan Zhou
| Challenge: | Large language models have shown potential in assisting scientific research, yet their ability to discover high-quality research hypotheses remains unexamined due to the lack of a dedicated benchmark. |
| Approach: | They propose a benchmark for evaluating large language models on a sufficient set of scientific discovery sub-tasks. |
| Outcome: | The proposed framework extracts critical components from papers across 12 disciplines with expert validation confirming its accuracy. |
Toward Automated Robustness Evaluation of Mathematical Reasoning (2026.findings-acl)
Copied to clipboard
Yutao Hou, Zeguan Xiao, Fei Yu, Yihan Jiang, Ma Shuguang, Zhaoqian Dai, Hailiang Huang, Yun Chen, Guanhua Chen
| Challenge: | Existing robustness evaluations rely on hand-crafted templates or a limited set of perturbation rules, resulting in model failure. |
| Approach: | They propose a framework inspired by software stress testing that generates adversarial variants via a multi-round rewrite-verify loop, ensuring semantic consistency while successfully inducing model failure. |
| Outcome: | The proposed framework generates adversarial variants dynamically for each LLM, minimizing the risk of data contamination. |
When Safety Alignment Fails to Generalize: Probing with Language Game Jailbreaks (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing safety alignment methods rely on fixed or narrow transformation schemes to generalize . existing methods based on fixed and narrow transformations are often inadequate . |
| Approach: | They propose a framework for discovering and refining language game-based jailbreaks to probe alignment generalization. |
| Outcome: | The proposed framework allows controlled exploration of alignment behavior across closely related linguistic variants. |
From Script to Stage: Automating Experimental Design for Social Simulations with LLMs (2026.findings-acl)
Copied to clipboard
| Challenge: | Xu et al., 2024): multi-agent simulations based on large language models are a new paradigm for social science research . traditional experimental design relies on interdisciplinary expertise and technical barriers . Xiaoping and Xin eli argue that LLM-driven agents are unreliable for rigorous experimental design due to hallucinations and limited verifiability. |
| Approach: | They propose a framework for multi-agent experiment design based on script generation . Script Composition, Script Finalization, and Actor Generation are the core phases of the framework . |
| Outcome: | The proposed framework lowers the barrier for social science experimental design and provides scientifically grounded decision support for policy-making. |
Insight Over Sight: Exploring the Vision-Knowledge Conflicts in Multimodal LLMs (2025.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to mitigating vision-knowledge conflict in Large Language Models (MLLMs) are not effective and can be further scaled. |
| Approach: | They propose a framework to generate inputs to simulate and evaluate vision-knowledge conflict in Multimodal Large Language Models (MLLMs) using original images and 1,122 high-quality question-answer pairs, they propose 'a diagnostic benchmark' |
| Outcome: | The proposed framework, benchmark, and analysis contribute to the understanding and mitigation of vision-knowledge conflicts in Multimodal Large Language Models (MLLMs). |
STARLING: Self-supervised Training of Text-based Reinforcement Learning Agent with Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing environments for text-based RL are domain-specific or time-consuming to generate and do not train the agents to master a specific set of skills. |
| Approach: | They propose an interactive environment for self-supervised RL that bootstraps the text-based RL agents with automatically generated games to boost their generalization capabilities. |
| Outcome: | The proposed environment bootstraps the agents with automatically generated games to boost their generalization capabilities to reach a goal of the target environment. |
ExpNote: Black-box Large Language Models are better Task Solvers with Experience Notebook (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have shown great power in solving various tasks but fail in many specific tasks. |
| Approach: | They propose a framework to help black-box LLMs better adapt to unfamiliar tasks by reflecting and noting experiences from training data and retrieving them from external memory during testing. |
| Outcome: | The proposed framework improves the performance of black-box Large Language Models on multiple tasks and demonstrates that it is a good choice for the future. |
Uncovering Factor-Level Preference to Improve Human-Model Alignment (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models exhibit tendencies that diverge from human preferences, such as favoring certain writing styles or producing overly verbose outputs. |
| Approach: | They propose a framework to uncover and measure factor-level preference alignment of humans and large language models (LLMs) |
| Outcome: | The proposed framework uncovers and measures factor-level preference alignment of humans and large language models. |
Understanding and Mitigating Overrefusal in LLMs from an Unveiling Perspective of Safety Decision Boundary (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) often refuse to answer legitimate queries, causing models to treat many reasonable prompts as potentially risky. |
| Approach: | They propose a framework that automatically generates and selects overrefusal prompts near the safety boundary. |
| Outcome: | The proposed framework identifies and curates boundary-aligned prompts, enabling more effective and targeted mitigation of overrefusal. |
Diagnosing Failures in Large Language Models’ Answers: Integrating Error Attribution into Evaluation Framework (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing evaluation models lack error attribution capability due to their proprietary nature. |
| Approach: | They propose a misattribution framework with 6 primary and 15 secondary categories to facilitate in-depth analysis. |
| Outcome: | The proposed framework is based on a dataset specifically designed for error attribution, along with the corresponding scores and feedback. |
CoreCodeBench: Decoupling Code Intelligence via Fine-Grained Repository-Level Tasks (2026.acl-long)
Copied to clipboard
Lingyue Fu, Hao Guan, Bolun Zhang, Haowei Yuan, Yaoming Zhu, Lin Qiu, ZongYu Wang, Xuezhi Cao, Xunliang Cai, Weiwen Liu, Weinan Zhang, Yong Yu
| Challenge: | Existing large language models for software engineering rely on coarse-grained pass rates obscuring specific cognitive bottlenecks. |
| Approach: | They propose a repository-level benchmark that dissects coding capabilities through atomized tasks. |
| Outcome: | The proposed framework achieves a 78.55% validity yield, surpassing the 31.7% retention rate of SWE-bench-Verified. |
COSMIC: Generalized Refusal Direction Identification in LLM Activations (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods depend on predefined refusal templates detectable in output tokens or manual review. |
| Approach: | They propose a framework that optimally identifies steering directions and target layers using cosine similarity, entirely independent of output text. |
| Outcome: | The proposed framework achieves comparable steering effectiveness without any prior knowledge or assumptions of a model’s refusal behavior such as the use of certain refusal tokens. |
PLAY2PROMPT: Zero-shot Tool Instruction Optimization for LLM Agents via Tool Play (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing solutions for large language models rely on manual rewriting or labeled data for validation . Existing approaches rely only on comprehensive tool documentation and in-context demonstrations . |
| Approach: | They propose a framework that "plays" with each tool to explore its input-output behaviors. |
| Outcome: | Experiments show that PLAY2PROMPT improves zero-shot tool performance across open and closed models. |
From Domains to Instances: Dual-Granularity Data Synthesis for LLM Unlearning (2026.findings-acl)
Copied to clipboard
Xiaoyu Xu, Minxin Du, Zitong LI, Zi Liang, Zhibiao Guo, Zhang Shiyu, Peizhao Hu, Qingqing Ye, Haibo Hu
| Challenge: | Currently, the evaluation of unlearning is limited due to the lack of granularity in the model. |
| Approach: | They propose a framework for synthesizing high-quality forget sets that exploits the target model per se to elicit data that matches its internal knowledge distribution through seed-guided and adversarial prompting. |
| Outcome: | The proposed framework achieves a superior balance of relevance, diversity, and efficiency across benchmarks. |
SOCIAL SCAFFOLDS: A Generalization Framework for Social Understanding Tasks (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing tools for social reasoning tasks lack explicit social cues, such as intents and implications. |
| Approach: | They propose a framework for facilitating generalization across social reasoning tasks by generating rationales that make social cues explicit. |
| Outcome: | The proposed framework improves cross-task performance and inferred speaker intent in 44% of cases. |
ASTRA: An Automated Framework for Strategy Discovery, Retrieval, and Evolution for Jailbreaking LLMs (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods lack the capability for continuous learning and self-evolution from interactions, limiting the diversity and adaptability of attack strategies. |
| Approach: | They propose an automated framework capable of discovering, retrieving, and evolving attack strategies. |
| Outcome: | The proposed framework outperforms existing baselines in a black-box setting. |