Junwei Bao

Yingyao Wang, Yongwei Zhou, Chaoqun Duan et al.

The semantic parsing-based method is an important research branch for knowledge-based question answering. It usually generates executable programs lean upon the question and then conduct them to reason answers over a knowledge base. Benefit from this inherent mechanism, it has advantages in the performance and the interpretability. However, traditional semantic parsing methods usually generate a complete program before executing it, which struggles with multi-hop question answering over heterogeneous knowledge. On one hand, generating a complete multi-hop program relies on multiple heterogeneous supporting facts, and it is difficult for generators to understand these facts simultaneously. On the other hand, this way ignores the semantic information of the intermediate answers at each hop, which is beneficial for subsequent generation. To alleviate these challenges, we propose a self-iterative framework for multi-hop program generation (HopPG) over heterogeneous knowledge, which leverages the previous execution results to retrieve supporting facts and generate subsequent programs hop by hop. We evaluate our model on MMQA-T^2, and the experimental results show that HopPG outperforms existing semantic-parsing-based baselines, especially on the multi-hop questions.

Haihui Pan, Junwei Bao, Hongfei Jiang et al.

While large language models have made significant progress in mathematical reasoning, they remain unreliable at judging the correctness of their own solutions. Existing approaches that equip models with self-verification typically treat solution generation and verification as two separate tasks, leading to substantially increased training time. In this paper, we propose FABSVer, which fuses these two tasks into a single generation pass, dramatically reducing training overhead while jointly optimizing both capabilities. We further identify a convergence bottleneck both theoretically and empirically: as training progresses, the reward reaches a plateau because the policy is constrained by a fixed reference model. To overcome this, we introduce Dynamic Reference Model Update (DRMU), which raises the reward ceiling and enables sustained reward growth. Extensive experiments on math benchmarks demonstrate that FABSVer achieves superior self-verification and reasoning performance across three model scales, while requiring only 51%--71% of the training time of existing methods. Analysis further reveals distinct learning phases in how models acquire self-verification, and that the gap between verify and answer rewards shrinks noticeably as model size increases.

Guanming Xiong, Junwei Bao, Hongfei Jiang et al.

This study explores text-to-SQL parsing by leveraging the powerful reasoning capabilities of large language models (LLMs). Despite recent advancements, existing LLM-based methods are still inefficient and struggle to handle cases with wide tables effectively. Furthermore, current interaction-based approaches either lack a step-by-step, interpretable SQL generation process or fail to provide a universally applicable interaction design. To address these challenges, we introduce Interactive-T2S, a framework that generates SQL queries through direct interactions with databases. This framework includes four general tools that facilitate proactive and efficient information retrieval by the LLM. Additionally, we have developed detailed exemplars to demonstrate the step-wise reasoning processes within our framework. Our approach achieves advanced performance on the Spider and BIRD datasets as well as their variants. Notably, we obtain state-of-the-art results on the BIRD leaderboard under the setting without oracle knowledge, demonstrating the effectiveness of our method.

Yuzhong Hong, Hanshan Zhang, Junwei Bao et al.

Since the debut of DPO, it has been shown that aligning a target LLM with human preferences via the KL-constrained RLHF loss is mathematically equivalent to a special kind of reward modeling task. Concretely, the task requires: 1) using the target LLM to parameterize the reward model, and 2) tuning the reward model so that it has a 1:1 linear relationship with the true reward. However, we identify a significant issue: the DPO loss might have multiple minimizers, of which only one satisfies the required linearity condition. The problem arises from a well-known issue of the underlying Bradley-Terry preference model: it does not always have a unique maximum likelihood estimator (MLE). Consequently,the minimizer of the RLHF loss might be unattainable because it is merely one among many minimizers of the DPO loss. As a better alternative, we propose an energy-based model (EBM) that always has a unique MLE, inherently satisfying the linearity requirement. To approximate the MLE in practice, we propose a contrastive loss named Energy Preference Alignment (EPA), wherein each positive sample is contrasted against one or more strong negatives as well as many free weak negatives. Theoretical properties of our EBM enable the approximation error of EPA to almost surely vanish when a sufficient number of negatives are used. Empirically, we demonstrate that EPA consistently delivers better performance on open benchmarks compared to DPO, thereby showing the superiority of our EBM.

Qiushi Wang, Yuchen Fan, Junwei Bao et al.

In recent years, Parameter-Efficient Fine-Tuning (PEFT) methods like Low-Rank Adaptation (LoRA) have significantly enhanced the adaptability of large-scale pre-trained models. Weight-Decomposed Low-Rank Adaptation (DoRA) improves upon LoRA by separating the magnitude and direction components of the weight matrix, leading to superior performance. However, DoRA's improvements are limited to the vertical dimension, resulting in an asymmetrical pattern between horizontal and vertical dimensions. This paper introduces BoRA, an innovative extension of LoRA and DoRA, characterized by symmetrical properties across horizontal and vertical dimensions. Our approach optimizes the weight matrix symmetrically by adjusting both column-wise and row-wise magnitudes. Extensive experiments demonstrate that BoRA surpasses state-of-the-art PEFT methods, including LoRA and DoRA, achieving superior results across various benchmarks.

Yuchen Fan, Yuzhong Hong, Qiushi Wang et al.

Alignment, endowing a pre-trained Large language model (LLM) with the ability to follow instructions, is crucial for its real-world applications. Conventional supervised fine-tuning (SFT) methods formalize it as causal language modeling typically with a cross-entropy objective, requiring a large amount of high-quality instruction-response pairs. However, the quality of widely used SFT datasets can not be guaranteed due to the high cost and intensive labor for the creation and maintenance in practice. To overcome the limitations associated with the quality of SFT datasets, we introduce a novel \textbf{p}reference-\textbf{o}riented supervised \textbf{f}ine-\textbf{t}uning approach, namely PoFT. The intuition is to boost SFT by imposing a particular preference: \textit{favoring the target model over aligned LLMs on the same SFT data.} This preference encourages the target model to predict a higher likelihood than that predicted by the aligned LLMs, incorporating assessment information on data quality (i.e., predicted likelihood by the aligned LLMs) into the training process. Extensive experiments are conducted, and the results validate the effectiveness of the proposed method. PoFT achieves stable and consistent improvements over the SFT baselines across different training datasets and base models. Moreover, we prove that PoFT can be integrated with existing SFT data filtering methods to achieve better performance, and further improved by following preference optimization procedures, such as DPO.

Jing Zhao, Junwei Bao, Yifan Wang et al.

Transformer-based pre-trained models, such as BERT, have achieved remarkable results on machine reading comprehension. However, due to the constraint of encoding length (e.g., 512 WordPiece tokens), a long document is usually split into multiple chunks that are independently read. It results in the reading field being limited to individual chunks without information collaboration for long document machine reading comprehension. To address this problem, we propose RoR, a read-over-read method, which expands the reading field from chunk to document. Specifically, RoR includes a chunk reader and a document reader. The former first predicts a set of regional answers for each chunk, which are then compacted into a highly-condensed version of the original document, guaranteeing to be encoded once. The latter further predicts the global answers from this condensed document. Eventually, a voting strategy is utilized to aggregate and rerank the regional and global answers for final prediction. Extensive experiments on two benchmarks QuAC and TriviaQA demonstrate the effectiveness of RoR for long document reading. Notably, RoR ranks 1st place on the QuAC leaderboard (https://quac.ai/) at the time of submission (May 17th, 2021).

Yongwei Zhou, Junwei Bao, Haipeng Sun et al.

Reasoning machine reading comprehension (R-MRC) aims to answer complex questions that require discrete reasoning based on text. To support discrete reasoning, evidence, typically the concise textual fragments that describe question-related facts, including topic entities and attribute values, are crucial clues from question to answer. However, previous end-to-end methods that achieve state-of-the-art performance rarely solve the problem by paying enough emphasis on the modeling of evidence, missing the opportunity to further improve the model's reasoning ability for R-MRC. To alleviate the above issue, in this paper, we propose an evidence-emphasized discrete reasoning approach (EviDR), in which sentence and clause level evidence is first detected based on distant supervision, and then used to drive a reasoning module implemented with a relational heterogeneous graph convolutional network to derive answers. Extensive experiments are conducted on DROP (discrete reasoning over paragraphs) dataset, and the results demonstrate the effectiveness of our proposed approach. In addition, qualitative analysis verifies the capability of the proposed evidence-emphasized discrete reasoning for R-MRC.

Yu Lu, Junwei Bao, Yan Song et al.

Existing conversational recommendation (CR) systems usually suffer from insufficient item information when conducted on short dialogue history and unfamiliar items. Incorporating external information (e.g., reviews) is a potential solution to alleviate this problem. Given that reviews often provide a rich and detailed user experience on different interests, they are potential ideal resources for providing high-quality recommendations within an informative conversation. In this paper, we design a novel end-to-end framework, namely, Review-augmented Conversational Recommender (RevCore), where reviews are seamlessly incorporated to enrich item information and assist in generating both coherent and informative responses. In detail, we extract sentiment-consistent reviews, perform review-enriched and entity-based recommendations for item suggestions, as well as use a review-attentive encoder-decoder for response generation. Experimental results demonstrate the superiority of our approach in yielding better performance on both recommendation and conversation responding.

Jing Zhao, Junwei Bao, Yifan Wang et al.

Keyphrases, that concisely summarize the high-level topics discussed in a document, can be categorized into present keyphrase which explicitly appears in the source text, and absent keyphrase which does not match any contiguous subsequence but is highly semantically related to the source. Most existing keyphrase generation approaches synchronously generate present and absent keyphrases without explicitly distinguishing these two categories. In this paper, a Select-Guide-Generate (SGG) approach is proposed to deal with present and absent keyphrase generation separately with different mechanisms. Specifically, SGG is a hierarchical neural network which consists of a pointing-based selector at low layer concentrated on present keyphrase generation, a selection-guided generator at high layer dedicated to absent keyphrase generation, and a guider in the middle to transfer information from selector to generator. Experimental results on four keyphrase generation benchmarks demonstrate the effectiveness of our model, which significantly outperforms the strong baselines for both present and absent keyphrases generation. Furthermore, we extend SGG to a title generation task which indicates its extensibility in natural language generation tasks.

Yingyao Wang, Junwei Bao, Guangyi Liu et al.

This paper aims to enhance the few-shot relation classification especially for sentences that jointly describe multiple relations. Due to the fact that some relations usually keep high co-occurrence in the same context, previous few-shot relation classifiers struggle to distinguish them with few annotated instances. To alleviate the above relation confusion problem, we propose CTEG, a model equipped with two mechanisms to learn to decouple these easily-confused relations. On the one hand, an Entity-Guided Attention (EGA) mechanism, which leverages the syntactic relations and relative positions between each word and the specified entity pair, is introduced to guide the attention to filter out information causing confusion. On the other hand, a Confusion-Aware Training (CAT) method is proposed to explicitly learn to distinguish relations by playing a pushing-away game between classifying a sentence into a true relation and its confusing relation. Extensive experiments are conducted on the FewRel dataset, and the results show that our proposed model achieves comparable and even much better results to strong baselines in terms of accuracy. Furthermore, the ablation test and case study verify the effectiveness of our proposed EGA and CAT, especially in addressing the relation confusion problem.

Junwei Bao, Duyu Tang, Nan Duan et al.

In this paper, we present a generative model to generate a natural language sentence describing a table region, e.g., a row. The model maps a row from a table to a continuous vector and then generates a natural language sentence by leveraging the semantics of a table. To deal with rare words appearing in a table, we develop a flexible copying mechanism that selectively replicates contents from the table in the output sequence. Extensive experiments demonstrate the accuracy of the model and the power of the copying mechanism. On two synthetic datasets, WIKIBIO and SIMPLEQUESTIONS, our model improves the current state-of-the-art BLEU-4 score from 34.70 to 40.26 and from 33.32 to 39.12, respectively. Furthermore, we introduce an open-domain dataset WIKITABLETEXT including 13,318 explanatory sentences for 4,962 tables. Our model achieves a BLEU-4 score of 38.23, which outperforms template based and language model based approaches.

Zhao Yan, Duyu Tang, Nan Duan et al.

Understanding the connections between unstructured text and semi-structured table is an important yet neglected problem in natural language processing. In this work, we focus on content-based table retrieval. Given a query, the task is to find the most relevant table from a collection of tables. Further progress towards improving this area requires powerful models of semantic matching and richer training and evaluation resources. To remedy this, we present a ranking based approach, and implement both carefully designed features and neural network architectures to measure the relevance between a query and the content of a table. Furthermore, we release an open-domain dataset that includes 21,113 web queries for 273,816 tables. We conduct comprehensive experiments on both real world and synthetic datasets. Results verify the effectiveness of our approach and present the challenges for this task.