Papers by Ben Zhang

20 papers
MSEarth: A Multimodal Benchmark for Earth Science Phenomenon Discovery with MLLMs (2026.acl-long)

Copied to clipboard

Challenge: Existing datasets often rely on synthetic data or figure-caption pairs, failing to capture the depth and complexity of geoscientific reasoning.
Approach: They propose a multimodal scientific dataset and benchmark curated from open-access publications.
Outcome: MSEarth features over 289K figures with captions enriched by contextual discussions and reasoning from original papers.
All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG (2026.acl-long)

Copied to clipboard

Challenge: Existing mRAG systems suffer from a language bias during reranking, systematically favoring English and the query’s native language.
Approach: They propose a language-agnostic utility-driven reranker alignment technique to mitigate language bias during re-ranking.
Outcome: The proposed approach mitigates language bias and consistently improves mRAG performance across languages.
Task-Oriented Dialogue as Dataflow Synthesis (2020.tacl-1)

Copied to clipboard

Challenge: Existing approaches to task-oriented dialogue represent dialogue state as a dataflow graph . microsoft's SMCalFlow dataset features complex dialogues about events, weather, places, and people .
Approach: They propose a dataflow graph-based dialogue agent that maps each user utterance to a program that extends this graph.
Outcome: The proposed framework improves representability and predictability in natural dialogues . it uses dataflow graphs and metacomputation to map user intents to a program .
Sub-Sentence Encoder: Contrastive Learning of Propositional Semantic Representations (2024.naacl-long)

Copied to clipboard

Challenge: Sentence embeddings are typically learned to recognize the semantic relation between two text inputs.
Approach: They introduce a contrastively-learned contextual embedding model for fine-grained semantic representation of text.
Outcome: The proposed model is able to produce contextual embeddings corresponding to different atomic propositions, i.e. semantic equivalence between propositions across different text sequences.
Reducing Gender Bias in Word-Level Language Models with a Gender-Equalizing Loss Function (P19-2)

Copied to clipboard

Challenge: Existing methods to reduce gender bias in natural language datasets are inadequate.
Approach: They propose a loss function modification approach which equalizes the probabilities of male and female words in the output.
Outcome: The proposed approach outperforms existing methods in several aspects, especially in reducing gender bias in occupation words.
On-Policy Self-Alignment with Fine-grained Knowledge Feedback for Hallucination Mitigation (2025.findings-acl)

Copied to clipboard

Challenge: Large language models exhibit behavior that deviates from the boundaries of their knowledge during response generation.
Approach: They propose a framework that allows large language models to explore their knowledge boundaries and self-correct generation behavior through fine-grained feedback signals.
Outcome: The proposed framework enables LLMs to explore their knowledge boundaries and self-correct generation behavior through fine-grained feedback signals.
RESIN: A Dockerized Schema-Guided Cross-document Cross-lingual Cross-media Information Extraction and Event Tracking System (2021.naacl-demos)

Copied to clipboard

Challenge: We present a new information extraction system that can construct temporal event graphs from news documents.
Approach: They propose a temporal event graph extraction system that can extract news documents . they extend the system from sentence-level event extraction to cross-document cross-media event extraction .
Outcome: The proposed system can extract temporal event graphs from news documents in multiple languages and multiple data modalities.
Enhancing LLM Capabilities Beyond Scaling Up (2024.emnlp-tutorials)

Copied to clipboard

Challenge: general-purpose large language models (LLMs) are expanding in scale and access to unpublic training data.
Approach: This tutorial aims to examine the capabilities of general-purpose large language models . authors discuss adaptation of LLMs to address conflicts, defense against attacks .
Outcome: This tutorial aims to examine the evolution of general-purpose large language models (LLMs) the authors argue that the evolution is dependent on the availability of training data and the scale of the models.
MoDULA: Mixture of Domain-Specific and Universal LoRA for Multi-Task Learning (2024.emnlp-main)

Copied to clipboard

Challenge: Recent advances in open-source Large Language Models (LLMs) have achieved notable successes in natural language processing.
Approach: They propose a Parameter Efficient Fine-Tuning paradigm for improved fine-tuning and parameter efficiency in multi-task learning.
Outcome: The proposed model outperforms existing methods on multi-task learning while reducing training costs by over 80% without losing general capability.
FeTaQA: Free-form Table Question Answering (2022.tacl-1)

Copied to clipboard

Challenge: Existing table-based question answering datasets lack advanced information-based questions that require reasoning and integration of information pieces retrieved from structured knowledge sources.
Approach: They propose a dataset with 10K Wikipedia-based table, question, free-form answer, supporting table cells pairs that can be used to generate an answer.
Outcome: The proposed dataset has 10K Wikipedia-based table, question, free-form answer, supporting table cells pairs.
LINC: A Neurosymbolic Approach for Logical Reasoning by Combining Language Models with First-Order Logic Provers (2023.emnlp-main)

Copied to clipboard

Challenge: Logical reasoning is an important task for artificial intelligence, says a new study . many prompting-based strategies to enable large language models fail in subtle and unpredictable ways.
Approach: They propose to reformulate logical reasoning tasks by leveraging large language models . they use a modular neurosymbolic programming approach to translate premises and conclusions from natural language to logic .
Outcome: The proposed approach outperforms open-source models on FOLIO and ProofWriter while showing distinct failure modes.
How can NLP Help Revitalize Endangered Languages? A Case Study and Roadmap for the Cherokee Language (2022.acl-long)

Copied to clipboard

Challenge: There are an estimated 6000 to 7000 spoken languages in the world, and at least 43% of them are endangered.
Approach: They propose three principles that may help NLP practitioners foster mutual understanding and collaboration with language communities and three ways in which NLP can potentially assist in language education.
Outcome: The proposed methods can be used to enrich Cherokee language resources with machine-in-the-loop processing and to provide language education.
Evaluating Models’ Local Decision Boundaries via Contrast Sets (2020.findings-emnlp)

Copied to clipboard

Challenge: Standard test sets for supervised learning evaluate in-distribution generalization but are misleading when a dataset has systematic gaps.
Approach: They propose a more rigorous annotation paradigm for NLP that helps to close systematic gaps in the test data.
Outcome: The proposed model performs significantly lower on contrast sets than on the original test sets—up to 25% in some cases.
OS-Sentinel: Towards Safety-Enhanced Mobile GUI Agents via Hybrid Validation in Realistic Workflows (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for detecting unsafe mobile GUI agents are underexplored.
Approach: They propose a mobile agent safety detection framework that integrates a formal verifier and a VLM-based contextual judge to detect system-level violations.
Outcome: The proposed framework achieves 10%–30% improvements over existing approaches across multiple metrics.
Scale Down to Speed Up: Dynamic Data Selection for Reinforcement Learning (2025.findings-emnlp)

Copied to clipboard

Challenge: Current approaches to Reinforcement Learning (RL) rely on massive static datasets, leading to computational inefficiency and redundant gradient updates.
Approach: They propose a data-centric RL framework that dynamically selects the most informative training samples to optimize RL for mathematical reasoning.
Outcome: The proposed framework achieves comparable performance to full-data training methods while requiring only 1.5K samples instead of 220K, reducing training time from 13 days to just 4 hours on 8A800 GPUs.
Cheems: A Practical Guidance for Building and Evaluating Chinese Reward Models from Scratch (2025.acl-long)

Copied to clipboard

Challenge: Existing Chinese resources are small in scale and limited to specific domains, making them insufficient for LLM post-training.
Approach: They propose a Chinese-annotated reward model and a preference dataset to address this gap . they evaluate Chinese RMs on CheemsBench and construct an RM that captures human preferences .
Outcome: The proposed RM achieves state-of-the-art performance on CheemsBench and CheeMePreference.
Reasoning in Token Economies: Budget-Aware Evaluation of LLM Reasoning Strategies (2024.emnlp-main)

Copied to clipboard

Challenge: Existing evaluations that focus on performance metrics miss a key factor: increased effectiveness due to additional compute.
Approach: They propose to incorporate the compute budget into evaluations to provide a more informative comparison that takes into account both performance metrics and computational cost.
Outcome: The proposed framework outperforms reasoning strategies when they use comparable compute resources.
Not quite Sherlock Holmes: Language model predictions do not reliably differentiate impossible from improbable events (2025.findings-acl)

Copied to clipboard

Challenge: Existing work has shown that language models can select the most likely or plausible of a set of possible events, but they are far from robust.
Approach: They focus on whether language models can select the most likely or plausible of a set of possibilities and compare them to a broader behavior that humans exhibit largely unconsciously.
Outcome: The proposed models perform worse than expected under certain conditions, compared with Llama 3, Gemma 2, and Mistral NeMo, and they are significantly more sensible than leaves.
Filter-then-Generate: Large Language Models with Structure-Text Adapter for Knowledge Graph Completion (2025.coling-main)

Copied to clipboard

Challenge: Empirical evidence suggests that LLMs perform worse than conventional KGC approaches.
Approach: They propose a filter-then-generate paradigm and a multiple-choice question format to harness the capability of LLMs while mitigating the issue casused by hallucinations.
Outcome: The proposed method achieves substantial performance gain compared to existing state-of-the-art methods.

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