Papers by Zhixuan Yang

5 papers
Perplexity-Aware Data Scaling Law: Perplexity Landscapes Predict Performance for Continual Pre-training (2026.acl-long)

Copied to clipboard

Challenge: Large language models (LLMs) have impressive capabilities across a wide range of domains, but their generalpurpose pre-training objectives often leave them illsuited for specialized applications such as healthcare.
Approach: They propose a perplexity-aware data scaling law that establishes a predictive relationship between the perplexities of domain-specific data and the test loss.
Outcome: Experiments on medical and general-domain benchmarks show that the proposed scaling law consistently identifies near-optimal training subsets with significantly reduced data consumption.
Adaptive Prompt Optimization for Open-Ended Tasks: Uncertainty Preference as a Secondary Signal (2026.findings-acl)

Copied to clipboard

Challenge: Recent training-free prompt optimizers treat performance as maximizing a single scalar score and ignore a second signal that the desired style is task dependent.
Approach: They propose a semantic-entropy-based method that uses task uncertainty to guide prompt optimization by selecting high-entropicy candidates for creative tasks and low-energetic candidates for conservative ones.
Outcome: The proposed method outperforms baselines on MT-Bench subsets and integrates easily into existing prompt optimizers.
ATGL: An Adaptive-Threshold Global Loss for Document-level Relation Extraction (2026.acl-long)

Copied to clipboard

Challenge: Document-level relation extraction (DocRE) aims to determine which relations hold between a given entity pair in a document.
Approach: They propose a document-level relation extraction paradigm that decouples existing losses into independent positive and negative losses, which interact solely with a shared threshold.
Outcome: The proposed model outperforms existing models on four datasets and achieves state-of-the-art results.
Inference-Time Language Model Alignment via Integrated Value Guidance (2024.findings-emnlp)

Copied to clipboard

Challenge: Large language models are fine-tuned to align with human preferences, but tuning large models is computationally intensive and complex.
Approach: They propose a method that uses implicit and explicit value functions to guide language model decoding at token and chunk-level respectively.
Outcome: The proposed method outperforms traditional methods and circumvents the complexities of fine-tuning.
DEBAR: Mitigating Contextual Bias in Cross-Document Relation Extraction via Dual-Stream Decoupling (2026.acl-long)

Copied to clipboard

Challenge: Existing methods focus on sentence-level or singledocument settings, resulting in one-sided relation transfer contextual bias and incomplete reasoning chains.
Approach: They propose a framework to explicitly decouple and preserve bidirectional bridge evidence and a dynamic loss optimization objective to separate head and tail contexts.
Outcome: The proposed framework decouples and preserves bidirectional bridge evidence while capturing global dependencies through iterative message passing.

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