Papers by Dongxu Zhang

9 papers
Fast and Effective On-Policy Distillation from Reasoning Prefixes (2026.findings-acl)

Copied to clipboard

Challenge: On-policy distillation (OPD) requires expensive on-the-fly sampling of the student policy during training, which substantially increases training cost.
Approach: They propose to use on-policy distillation to sample trajectories from student model . they propose to terminate the sampling early during distillation .
Outcome: The proposed method matches the performance of full OPD in long reasoning outputs while reducing training FLOP by 2x–40x.
A Distant Supervision Corpus for Extracting Biomedical Relationships Between Chemicals, Diseases and Genes (2022.lrec-1)

Copied to clipboard

Challenge: Biomedical researchers have used manual curation to extract biomedical interactions from research texts to improve coverage.
Approach: They propose a new dataset for training and evaluating multi-class multi-label biomedical relation extraction models using human annotations and the CTD database.
Outcome: The proposed dataset is substantially larger and cleaner than existing datasets and includes annotations linking mentions to their entities.
OpenKI: Integrating Open Information Extraction and Knowledge Bases with Relation Inference (N19-1)

Copied to clipboard

Challenge: Existing methods for knowledge extraction and alignment are limited in quality and performance.
Approach: They propose to integrate OpenIE extractions in the form of (subject, predicate, object) triples with Knowledge Bases (KB)
Outcome: The proposed method improves state-of-the-art for OpenIE extractions and boosts performance on OpenIE from semi-structured data.
SyntaxSQLNet: Syntax Tree Networks for Complex and Cross-Domain Text-to-SQL Task (D18-1)

Copied to clipboard

Challenge: Existing studies in text-to-SQL do not require generating complex SQL queries with multiple clauses or sub-queries.
Approach: They propose a syntax tree network to address the complex text-to-SQL generation task.
Outcome: The proposed model outperforms the current state-of-the-art model by 9.5% on a large text-to-SQL corpus.
Enhanced Distant Supervision with State-Change Information for Relation Extraction (2022.lrec-1)

Copied to clipboard

Challenge: Existing methods for enhancing distant supervision with state-change information for relation extraction are limited.
Approach: They propose a method for enhancing distant supervision with state-change information for relation extraction by adding temporal information to a curation dataset.
Outcome: The proposed method reduces noise when used for static relation extraction and can be used to train a relation-extraction system that detects a change of state in relations.
Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task (D18-1)

Copied to clipboard

Challenge: Existing datasets for semantic parsing are too small in terms of number of programs for training modern data-intensive models.
Approach: They propose a large-scale complex and cross-domain semantic parsing task for a database . they use a dataset with 10,181 questions and 5,693 unique complex SQL queries .
Outcome: The proposed task is different from previous tasks because it uses the same database and program . the best model achieves only 9.7% exact matching accuracy on a database split setting.
Event-Event Relation Extraction using Probabilistic Box Embedding (2022.acl-short)

Copied to clipboard

Challenge: Existing frameworks of event relation extraction do not guarantee coherence between different relation types, such as anti-symmetry.
Approach: They propose to modify existing ERE framework to guarantee coherence by representing each event as a box representation without applying explicit constraints.
Outcome: The proposed model shows stronger conjunctive constraint satisfaction compared to previous models with constraint injection.
Chain-of-Thought Compression Should Not Be Blind: V-Skip for Efficient Multimodal Reasoning via Dual-Path Anchoring (2026.acl-long)

Copied to clipboard

Challenge: Existing efforts to mitigate this via token compression fail due to its autoregressive nature . linguistically redundant tokens are erroneously pruned, leading to hallucinations .
Approach: They propose a method that reformulates token pruning as a Visual-Anchored Information Bottleneck (VA-IB) optimization problem.
Outcome: Experiments on Qwen2-VL and Llama-3.2 families show that the proposed model achieves a speedup with negligible accuracy loss.
Enhancing Hallucination Detection through Perturbation-Based Synthetic Data Generation in System Responses (2024.findings-acl)

Copied to clipboard

Challenge: Existing methods for hallucination detection are expensive and outdated . despite the popularity of LLMs, the issue of hallucinosity poses significant concerns for downstream users.
Approach: They propose an approach that automatically generates both faithful and hallucinated outputs by rewriting system responses.
Outcome: The proposed model outperforms state-of-the-art zero-shot detectors and existing synthetic generation methods in accuracy and latency.

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