Reinforcement learning from human feedback serves as a crucial bridge, aligning large language models with human and societal values. This alignment requires a vast corpus of human feedback to learn a reward model, which is subsequently used to finetune language models. However, we have identified that the reward model often finds shortcuts to bypass its intended objectives, misleadingly assuming that humans prefer longer responses. The emergence of length bias often induces the model to favor longer outputs, yet it doesn’t equate to an increase in helpful information within these outputs. In this paper, we propose an innovative solution, applying the Product-of-Experts (PoE) technique to separate reward modeling from the influence of sequence length. In our framework, the main expert concentrates on understanding human intents, while the biased expert targets the identification and capture of length bias. To further enhance the learning of bias, we introduce perturbations into the bias-focused expert, disrupting the flow of semantic information. Experimental results validate the effectiveness of our approach, indicating that language model performance is improved, irrespective of sequence length.

Semantic matching is a mainstream paradigm of zero-shot relation extraction, which matches a given input with a corresponding label description. The entities in the input should exactly match their hypernyms in the description, while the irrelevant contexts should be ignored when matching. However, general matching methods lack explicit modeling of the above matching pattern. In this work, we propose a fine-grained semantic matching method tailored for zero-shot relation extraction. Guided by the above matching pattern, we decompose the sentence-level similarity score into the entity matching score and context matching score. Considering that not all contextual words contribute equally to the relation semantics, we design a context distillation module to reduce the negative impact of irrelevant components on context matching. Experimental results show that our method achieves higher matching accuracy and more than 10 times faster inference speed, compared with the state-of-the-art methods.

The existing supervised relation extraction methods have achieved impressive performance in a closed-set setting, in which the relations remain the same during both training and testing. In a more realistic open-set setting, unknown relations may appear in the test set. Due to the lack of supervision signals from unknown relations, a well-performing closed-set relation extractor can still confidently misclassify them into known relations. In this paper, we propose an unknown-aware training method, regularizing the model by dynamically synthesizing negative instances that can provide the missing supervision signals. Inspired by text adversarial attack, We adaptively apply small but critical perturbations to original training data,synthesizing difficult enough negative instances that are mistaken by the model as known relations, thus facilitating a compact decision boundary. Experimental results show that our method achieves SOTA unknown relation detection without compromising the classification of known relations.

The introduction of multimodal information and pretraining technique significantly improves entity recognition from visually-rich documents. However, most of the existing methods pay unnecessary attention to irrelevant regions of the current document while ignoring the potentially valuable information in related documents. To deal with this problem, this work proposes a cross-document semantic enhancement method, which consists of two modules: 1) To prevent distractions from irrelevant regions in the current document, we design a learnable attention mask mechanism, which is used to adaptively filter redundant information in the current document. 2) To further enrich the entity-related context, we propose a cross-document information awareness technique, which enables the model to collect more evidence across documents to assist in prediction. The experimental results on two documents understanding benchmarks covering eight languages demonstrate that our method outperforms the SOTA methods.