Reinforcement learning (RL) has been widely adopted to enhance the performance of large language models (LLMs) on Text-to-SQL tasks. However, existing methods often rely on execution-based or LLM-based Bradley–Terry reward models. The former suffers from high execution latency caused by repeated database calls, whereas the latter imposes substantial GPU memory overhead, both of which significantly hinder the efficiency and scalability of RL pipelines. To this end, we propose a novel reward model framework for RL-based Text-to-SQL named Graph-Reward-SQL, which employs the GMNScore outcome reward model. We leverage SQL graph representations to provide accurate reward signals while significantly reducing time cost and GPU memory usage. Building on this foundation, we further introduce StepRTM, a stepwise reward model that provides intermediate supervision over Common Table Expression (CTE) subqueries. This encourages both functional correctness and readability of SQL. Extensive comparative and ablation experiments on standard benchmarks, including Spider and BIRD, demonstrate that our method consistently outperforms existing reward models.

Competitive programming, due to its high reasoning difficulty and precise correctness feedback, has become a key task for both training and evaluating the reasoning capabilities of large language models (LLMs). However, while a large amount of public problem data, such as problem statements and solutions, is available, the test cases of these problems are often difficult to obtain. Therefore, test case generation is a necessary task for building large-scale datasets, and the quality of the test cases directly determines the accuracy of the evaluation. In this paper, we introduce an LLM-based agent system that creates high-quality test cases for competitive programming problems. We apply this system to the CodeContests dataset and propose a new version with improved test cases, named CodeContests+. We evaluated the quality of test cases in CodeContests+. First, we used 1.72 million submissions with pass/fail labels to examine the accuracy of these test cases in evaluation. The results indicated that CodeContests+ achieves significantly higher accuracy than CodeContests, particularly with a notably higher True Positive Rate (TPR). Subsequently, our experiments in LLM Reinforcement Learning (RL) further confirmed that improvements in test case quality yield considerable advantages for RL.

Execution Accuracy and Exact Set Match are two predominant metrics for evaluating the functional correctness of SQL queries in modern Text-to-SQL tasks. However, both metrics have notable limitations: Exact Set Match fails when queries are functionally equivalent but syntactically different, while Execution Accuracy is prone to false positives due to inadequately prepared test databases, which can be costly to create, particularly in large-scale industrial applications. To overcome these challenges, we propose a novel graph-based metric, FuncEvalGMN, that effectively overcomes the deficiencies of the aforementioned metric designs. Our method utilizes a relational operator tree (ROT), referred to as RelNode, to extract rich semantic information from the logical execution plan of SQL queries, and embed it into a graph. We then train a graph neural network (GNN) to perform graph matching on pairs of SQL queries through graph contrastive learning. FuncEvalGMN offers two highly desired advantages: (i) it requires only the database schema to derive logical execution plans, eliminating the need for extensive test database preparation, and (ii) it demonstrates strong generalization capabilities on unseen datasets. These properties highlight FuncEvalGMN’s robustness as a reliable metric for assessing functional correctness across a wide range of Text-to-SQL applications.

Argumentation is a fundamental human activity that involves reasoning and persuasion, which also serves as the basis for the development of AI systems capable of complex reasoning. In NLP, to better understand human argumentation, argument structure analysis aims to identify argument components, such as claims and premises, and their relations from free text. It encompasses a variety of divergent tasks, such as end-to-end argument mining, argument pair extraction, and argument quadruplet extraction. Existing methods are usually tailored to only one specific argument structure analysis task, overlooking the inherent connections among different tasks. We observe that the fundamental goal of these tasks is similar: identifying argument components and their interrelations. Motivated by this, we present a unified generative framework for argument structure analysis (UniASA). It can uniformly address multiple argument structure analysis tasks in a sequence-to-sequence manner. Further, we enhance UniASA with a multi-view learning strategy based on subtask decomposition. We conduct experiments on seven datasets across three tasks. The results indicate that UniASA can address these tasks uniformly and achieve performance that is either superior to or comparable with the previous state-of-the-art methods. Also, we show that UniASA can be effectively integrated with large language models, such as Llama, through fine-tuning or in-context learning.

In this paper, we describe our submission for the NLI4CT 2024 shared task on robust Natural Language Inference over clinical trial reports. Our system is an ensemble of nine diverse models which we aggregate via majority voting. The models use a large spectrum of different approaches ranging from a straightforward Convolutional Neural Network over fine-tuned Large Language Models to few-shot-prompted language models using chain-of-thought reasoning.Surprisingly, we find that some individual ensemble members are not only more accurate than the final ensemble model but also more robust.

Argumentation mining (AM) aims to detect the arguments and their inherent relations from argumentative textual compositions. Generally, AM comprises three key challenging subtasks, including argument component type classification (ACTC), argumentative relation identification (ARI), and argumentative relation type classification (ARTC). Prior methods are afflicted by a sequential feature decoding paradigm, wherein they initially address the features of argumentation components (ACs) for the task of ACTC. Then, these features are amalgamated in pairs to tackle the task of ARI. Finally, the AC pairs and ascertained pertinent relations are employed for ARTC. However, the explicit and comprehensive inter-relationship among the three subtasks is neglected. In this paper, we propose a novel method PITA for PromptIng Task interAction to model the inter-relationships among the three subtasks within a generative framework. Specifically, we employ a dynamic prompt template to indicate all ACs and AC pairs in the three subtasks. Then, from a multi-relational perspective, we construct an undirected heterogeneous graph to capture the various relationships within and between ACs and AC pairs. We apply the Relational Graph Convolutional Network (RGCN) on the graph and inject the task interaction information into the soft prompts with continuous representations. PITA jointly decodes all ACs and AC pairs using the prompt template with task interaction information, which thus explicitly and comprehensively harmonizes the information propagation across the three subtasks. Extensive experiments show PITA achieves state-of-the-art performances on two AM benchmarks.

Extracting fine-grained structural information between argumentation component (AC) pairs is essential for argumentation relation classification (ARC). However, most previous studies attempt to model the relationship between AC pairs using AC level similarity or semantically relevant features. They ignore the complex interaction between AC pairs and cannot effectively reason the argumentation relation deeply.Therefore, in this paper, we propose a novel dual prior graph neural network (DPGNN) to jointly explore the probing knowledge derived from pre-trained language models (PLMs) and the syntactical information for comprehensively modeling the relationship between AC pairs. Specifically, we construct a probing graph by using probing knowledge derived from PLMs to recognize and align the relational information within and across the argumentation components. In addition, we propose a mutual dependency graph for the AC pair to reason the fine-grained syntactic structural information, in which the syntactical correlation between words is set by the dependency information within AC and mutual attention mechanism across ACs. The knowledge learned from the probing graph and the dependency graph are combined to comprehensively capture the aligned relationships of AC pairs for improving the results of ARC. Experimental results on three public datasets show that DPGNN outperforms the state-of-the-art baselines by a noticeable margin.

Argument mining (AM) is a challenging task as it requires recognizing the complex argumentation structures involving multiple subtasks.To handle all subtasks of AM in an end-to-end fashion, previous works generally transform AM into a dependency parsing task.However, such methods largely require complex pre- and post-processing to realize the task transformation.In this paper, we investigate the end-to-end AM task from a novel perspective by proposing a generative framework, in which the expected outputs of AM are framed as a simple target sequence. Then, we employ a pre-trained sequence-to-sequence language model with a constrained pointer mechanism (CPM) to model the clues for all the subtasks of AM in the light of the target sequence. Furthermore, we devise a reconstructed positional encoding (RPE) to alleviate the order biases induced by the autoregressive generation paradigm.Experimental results show that our proposed framework achieves new state-of-the-art performance on two AM benchmarks.

Due to the increasing use of service chatbots in E-commerce platforms in recent years, customer satisfaction prediction (CSP) is gaining more and more attention. CSP is dedicated to evaluating subjective customer satisfaction in conversational service and thus helps improve customer service experience. However, previous methods focus on modeling customer-chatbot interaction across different turns, which are hard to represent the important dynamic satisfaction states throughout the customer journey. In this work, we investigate the problem of satisfaction states tracking and its effects on CSP in E-commerce service chatbots. To this end, we propose a dialogue-level classification model named DialogueCSP to track satisfaction states for CSP. In particular, we explore a novel two-step interaction module to represent the dynamic satisfaction states at each turn. In order to capture dialogue-level satisfaction states for CSP, we further introduce dialogue-aware attentions to integrate historical informative cues into the interaction module. To evaluate the proposed approach, we also build a Chinese E-commerce dataset for CSP. Experiment results demonstrate that our model significantly outperforms multiple baselines, illustrating the benefits of satisfaction states tracking on CSP.

Emotion recognition in multi-party conversation (ERMC) is becoming increasingly popular as an emerging research topic in natural language processing. Prior research focuses on exploring sequential information but ignores the discourse structures of conversations. In this paper, we investigate the importance of discourse structures in handling informative contextual cues and speaker-specific features for ERMC. To this end, we propose a discourse-aware graph neural network (ERMC-DisGCN) for ERMC. In particular, we design a relational convolution to lever the self-speaker dependency of interlocutors to propagate contextual information. Furthermore, we exploit a gated convolution to select more informative cues for ERMC from dependent utterances. The experimental results show our method outperforms multiple baselines, illustrating that discourse structures are of great value to ERMC.