arXiv subjects

Chenxu Yang

Publications and source records attributed to Chenxu Yang.

30 records · Page 2Linked to original sources

arXiv2025

Breaking the Trade-Off Between Faithfulness and Expressiveness for Large Language Models

Grounding responses in external knowledge represents an effective strategy for mitigating hallucinations in Large Language Models (LLMs). However, current LLMs struggle to seamlessly integrate knowledge while simultaneously maintaining faithfulness (or fidelity) and expressiveness, capabilities that humans naturally possess. This limitation results in outputs that either lack support from external knowledge, thereby compromising faithfulness, or appear overly verbose and unnatural, thus sacrificing expressiveness. In this work, to break the trade-off between faithfulness and expressiveness, we propose Collaborative Decoding (CoDe), a novel approach that dynamically integrates output probabilities generated with and without external knowledge. This integration is guided by distribution divergence and model confidence, enabling the selective activation of relevant and reliable expressions from the model's internal parameters. Furthermore, we introduce a knowledge-aware reranking mechanism that prevents over-reliance on prior parametric knowledge while ensuring proper utilization of provided external information. Through comprehensive experiments, our plug-and-play CoDe framework demonstrates superior performance in enhancing faithfulness without compromising expressiveness across diverse LLMs and evaluation metrics, validating both its effectiveness and generalizability.

cs.CL↗

arXiv2025

A Factuality and Diversity Reconciled Decoding Method for Knowledge-Grounded Dialogue Generation

Grounding external knowledge can enhance the factuality of responses in dialogue generation. However, excessive emphasis on it might result in the lack of engaging and diverse expressions. Through the introduction of randomness in sampling, current approaches can increase the diversity. Nevertheless, such sampling method could undermine the factuality in dialogue generation. In this study, to discover a solution for advancing creativity without relying on questionable randomness and to subtly reconcile the factuality and diversity within the source-grounded paradigm, a novel method named DoGe is proposed. DoGe can dynamically alternate between the utilization of internal parameter knowledge and external source knowledge based on the model's factual confidence. Extensive experiments on three widely-used datasets show that DoGe can not only enhance response diversity but also maintain factuality, and it significantly surpasses other various decoding strategy baselines.

cs.CL↗

arXiv2025

Orthogonal Finetuning for Direct Preference Optimization

DPO is an effective preference optimization algorithm. However, the DPO-tuned models tend to overfit on the dispreferred samples, manifested as overly long generations lacking diversity. While recent regularization approaches have endeavored to alleviate this issue by modifying the objective function, they achieved that at the cost of alignment performance degradation. In this paper, we innovatively incorporate regularization from the perspective of weight updating to curb alignment overfitting. Through the pilot experiment, we discovered that there exists a positive correlation between overfitting and the hyperspherical energy fluctuation. Hence, we introduce orthogonal finetuning for DPO via a weight-Rotated Preference Optimization (RoPO) method, which merely conducts rotational and magnitude-stretching updates on the weight parameters to maintain the hyperspherical energy invariant, thereby preserving the knowledge encoded in the angle between neurons. Extensive experiments demonstrate that our model aligns perfectly with human preferences while retaining the original expressive capacity using only 0.0086% of the trainable parameters, suggesting an effective regularization against overfitting. Specifically, RoPO outperforms DPO by up to 10 points on MT-Bench and by up to 2.8 points on AlpacaEval 2, while enhancing the generation diversity by an average of 6 points.

cs.CL↗

arXiv2025

Weights-Rotated Preference Optimization for Large Language Models

Despite the efficacy of Direct Preference Optimization (DPO) in aligning Large Language Models (LLMs), reward hacking remains a pivotal challenge. This issue emerges when LLMs excessively reduce the probability of rejected completions to achieve high rewards, without genuinely meeting their intended goals. As a result, this leads to overly lengthy generation lacking diversity, as well as catastrophic forgetting of knowledge. We investigate the underlying reason behind this issue, which is representation redundancy caused by neuron collapse in the parameter space. Hence, we propose a novel Weights-Rotated Preference Optimization (RoPO) algorithm, which implicitly constrains the output layer logits with the KL divergence inherited from DPO and explicitly constrains the intermediate hidden states by fine-tuning on a multi-granularity orthogonal matrix. This design prevents the policy model from deviating too far from the reference model, thereby retaining the knowledge and expressive capabilities acquired during pre-training and SFT stages. Our RoPO achieves up to a 3.27-point improvement on AlpacaEval 2, and surpasses the best baseline by 6.2 to 7.5 points on MT-Bench with merely 0.015% of the trainable parameters, demonstrating its effectiveness in alleviating the reward hacking problem of DPO.

cs.CL↗

arXiv2025

Constructing disjoint Steiner trees in Sierpiński graphs

Let $G$ be a graph and $S\subseteq V(G)$ with $|S|\geq 2$. Then the trees $T_1, T_2, \cdots, T_\ell$ in $G$ are \emph{internally disjoint Steiner trees} connecting $S$ (or $S$-Steiner trees) if $E(T_i) \cap E(T_j )=\emptyset$ and $V(T_i)\cap V(T_j)=S$ for every pair of distinct integers $i,j$, $1 \leq i, j \leq \ell$. Similarly, if we only have the condition $E(T_i) \cap E(T_j )=\emptyset$ but without the condition $V(T_i)\cap V(T_j)=S$, then they are \emph{edge-disjoint Steiner trees}. The \emph{generalized $k$-connectivity}, denoted by $κ_k(G)$, of a graph $G$, is defined as $κ_k(G)=\min\{κ_G(S)|S \subseteq V(G) \ \textrm{and} \ |S|=k \}$, where $κ_G(S)$ is the maximum number of internally disjoint $S$-Steiner trees. The \emph{generalized local edge-connectivity} $λ_{G}(S)$ is the maximum number of edge-disjoint Steiner trees connecting $S$ in $G$. The {\it generalized $k$-edge-connectivity} $λ_k(G)$ of $G$ is defined as $λ_k(G)=\min\{λ_{G}(S)\,|\,S\subseteq V(G) \ and \ |S|=k\}$. These measures are generalizations of the concepts of connectivity and edge-connectivity, and they and can be used as measures of vulnerability of networks. It is, in general, difficult to compute these generalized connectivities. However, there are precise results for some special classes of graphs. In this paper, we obtain the exact value of $λ_{k}(S(n,\ell))$ for $3\leq k\leq \ell^n$, and the exact value of $κ_{k}(S(n,\ell))$ for $3\leq k\leq \ell$, where $S(n, \ell)$ is the Sierpiński graphs with order $\ell^n$. As a direct consequence, these graphs provide additional interesting examples when $λ_{k}(S(n,\ell))=κ_{k}(S(n,\ell))$. We also study the some network properties of Sierpiński graphs.

math.CO↗

arXiv2025

S-GRPO: Early Exit via Reinforcement Learning in Reasoning Models

As Test-Time Scaling emerges as an active research focus in the large language model community, advanced post-training methods increasingly emphasize extending chain-of-thought (CoT) generation length, thereby enhancing reasoning capabilities to approach Deepseek R1-like reasoning models. However, recent studies reveal that reasoning models (even Qwen3) consistently exhibit excessive thought redundancy in CoT generation. This overthinking issue arises from the inherent limitations of conventional outcome-reward reinforcement learning, which systematically overlooks the regulation of intermediate reasoning processes. This paper introduces Serial-Group Decaying-Reward Policy Optimization (S-GRPO), a novel reinforcement learning paradigm that enables models to implicitly evaluate the sufficiency of intermediate reasoning steps, thereby facilitating early exit in CoT generation. Unlike GRPO, which samples multiple possible reasoning paths in parallel (parallel group), S-GRPO only samples one reasoning path and serially selects multiple temporal positions from the path to exit thinking and directly generate answers (serial group). For correct answers within a serial group, rewards gradually decrease based on the exit positions along the reasoning path from front to back. This design encourages the model to produce more accurate and concise thoughts, while also incentivizing early thinking termination when appropriate. Empirical evaluations demonstrate that S-GRPO is compatible with state-of-the-art reasoning models, including Qwen3 and Deepseek-distill. Across diverse benchmarks such as GSM8K, AIME 2024, AMC 2023, MATH-500, and GPQA Diamond, S-GRPO achieves a substantial reduction in sequence length (35.4% - 61.1%) while simultaneously improving accuracy (absolute 0.72% - 6.08%).

cs.AI↗

arXiv2025

Gallai-Schur Triples and Related Problems

Schur's Theorem states that, for any $r \in \mathbb{Z}^+$, there exists a minimum integer $S(r)$ such that every $r$-coloring of $\{1,2,\dots,S(r)\}$ admits a monochromatic solution to $x+y=z$. Recently, Budden determined the related Gallai-Schur numbers; that is, he determined the minimum integer $GS(r)$ such that every $r$-coloring of $\{1,2,\dots,GS(r)\}$ admits either a rainbow or monochromatic solution to $x+y=z$. In this article we consider problems that have been solved in the monochromatic setting under a monochromatic-rainbow paradigm. In particular, we investigate Gallai-Schur numbers when $x \neq y$, we consider $x+y+b=z$ and $x+y<z$, and we investigate the asymptotic minimum number of rainbow and monochromatic solutions to $x+y=z$ and $x+y<z$.

math.CO↗

arXiv2025

Sibyl: Empowering Empathetic Dialogue Generation in Large Language Models via Sensible and Visionary Commonsense Inference

Recently, there has been a heightened interest in building chatbots based on Large Language Models (LLMs) to emulate human-like qualities in multi-turn conversations. Despite having access to commonsense knowledge to better understand the psychological aspects and causality of dialogue context, even these powerful LLMs struggle to achieve the goals of empathy and emotional support. Current commonsense knowledge derived from dialogue contexts is inherently limited and often fails to adequately anticipate the future course of a dialogue. This lack of foresight can mislead LLMs and hinder their ability to provide effective support. In response to this challenge, we present an innovative framework named Sensible and Visionary Commonsense Knowledge (Sibyl). Designed to concentrate on the immediately succeeding dialogue, this paradigm equips LLMs with the capability to uncover the implicit requirements of the conversation, aiming to elicit more empathetic responses. Experimental results demonstrate that incorporating our paradigm for acquiring commonsense knowledge into LLMs comprehensively enhances the quality of their responses.

cs.CL↗

arXiv2024

A Multi-Task Role-Playing Agent Capable of Imitating Character Linguistic Styles

The advent of large language models (LLMs) has significantly propelled the advancement of Role-Playing Agents (RPAs). However, current Role-Playing Agents predominantly focus on mimicking a character's fundamental attributes while neglecting the replication of linguistic style, and they are incapable of effectively replicating characters when performing tasks beyond multi-turn dialogues, which results in generated responses that lack authenticity. The reason current RPAs lack this capability is due to the nature of existing character datasets, which lack collections of character quotations and are limited to multi-turn dialogue tasks, constraining the RPA's performance across other task domains and failing to mimic a character's linguistic style. To address this gap, we developed a multi-task role-playing dataset named MRstyle, which encompasses a substantial number of real individuals along with their quotations and covers seven different tasks. On this basis, we develop StyleRPA, a Multi-Task Role-Playing Agent (MRPA) that significantly outperforms recent open-source LLMs and RPAs baselines on 7 tasks including Dialogue, Dictionary, Composition, Story Generation, Product Description, Music Commentary, and Open Question Answering. The code and data will be released.

cs.CL↗

arXiv2024

Perturbation results for distance-edge-monitoring numbers

Foucaud et al. recently introduced and initiated the study of a new graph-theoretic concept in the area of network monitoring. Given a graph $G=(V(G), E(G))$, a set $M \subseteq V(G)$ is a distance-edge-monitoring set if for every edge $e \in E(G)$, there is a vertex $x \in M$ and a vertex $y \in V(G)$ such that the edge $e$ belongs to all shortest paths between $x$ and $y$. The smallest size of such a set in $G$ is denoted by $\operatorname{dem}(G)$. Denoted by $G-e$ (resp. $G \backslash u$) the subgraph of $G$ obtained by removing the edge $e$ from $G$ (resp. a vertex $u$ together with all its incident edges from $G$). In this paper, we first show that $\operatorname{dem}(G-e)- \operatorname{dem}(G)\leq 2$ for any graph $G$ and edge $e \in E(G)$. Moreover, the bound is sharp. Next, we construct two graphs $G$ and $H$ to show that $\operatorname{dem}(G)-\operatorname{dem}(G\setminus u)$ and $\operatorname{dem}(H\setminus v)-\operatorname{dem}(H)$ can be arbitrarily large, where $u \in V(G)$ and $v \in V(H)$. We also study the relation between $\operatorname{dem}(H)$ and $\operatorname{dem}(G)$, where $H$ is a subgraph of $G$. In the end, we give an algorithm to judge whether the distance-edge monitoring set still remain in the resulting graph when any edge of the graph $G$ is deleted.

cs.DM↗

arXiv2023

Multi-level Adaptive Contrastive Learning for Knowledge Internalization in Dialogue Generation

Knowledge-grounded dialogue generation aims to mitigate the issue of text degeneration by incorporating external knowledge to supplement the context. However, the model often fails to internalize this information into responses in a human-like manner. Instead, it simply inserts segments of the provided knowledge into generic responses. As a result, the generated responses tend to be tedious, incoherent, and in lack of interactivity which means the degeneration problem is still unsolved. In this work, we first find that such copying-style degeneration is primarily due to the weak likelihood objective, which allows the model to "cheat" the objective by merely duplicating knowledge segments in a superficial pattern matching based on overlap. To overcome this challenge, we then propose a Multi-level Adaptive Contrastive Learning (MACL) framework that dynamically samples negative examples and subsequently penalizes degeneration behaviors at both the token-level and sequence-level. Extensive experiments on the WoW dataset demonstrate the effectiveness of our approach across various pre-trained models.

cs.CL↗

arXiv2022

Empathetic Dialogue Generation via Sensitive Emotion Recognition and Sensible Knowledge Selection

Empathy, which is widely used in psychological counselling, is a key trait of everyday human conversations. Equipped with commonsense knowledge, current approaches to empathetic response generation focus on capturing implicit emotion within dialogue context, where the emotions are treated as a static variable throughout the conversations. However, emotions change dynamically between utterances, which makes previous works difficult to perceive the emotion flow and predict the correct emotion of the target response, leading to inappropriate response. Furthermore, simply importing commonsense knowledge without harmonization may trigger the conflicts between knowledge and emotion, which confuse the model to choose incorrect information to guide the generation process. To address the above problems, we propose a Serial Encoding and Emotion-Knowledge interaction (SEEK) method for empathetic dialogue generation. We use a fine-grained encoding strategy which is more sensitive to the emotion dynamics (emotion flow) in the conversations to predict the emotion-intent characteristic of response. Besides, we design a novel framework to model the interaction between knowledge and emotion to generate more sensible response. Extensive experiments on EmpatheticDialogues demonstrate that SEEK outperforms the strong baselines in both automatic and manual evaluations.

cs.CL↗