arXiv subjects

Liting Zhang

Publications and source records attributed to Liting Zhang.

7 recordsLinked to original sources

arXiv2026

Is More Privileged Information Better? From Solution Traces to Problem-Solving Structure in Self-Distilled Reasoning

On-policy self-distillation (OPSD) improves reasoning by using a privileged view of a model conditioned on reference solutions to supervise a student view that observes only the question. However, the teacher-provided token-level targets may depend on reference-specific information unavailable at inference time. We propose Problem-Space-Guided OPSD (PS-OPSD), which replaces the complete solution with trajectory-grounded guidance describing the initial state, goal conditions, constraints, and a selected state-transition path. The student rollout and OPSD objective remain unchanged. Across three mathematical reasoning benchmarks and model scales ranging from 1.7B to 8B, PS-OPSD achieves the highest aggregate question-only accuracy among the compared methods. Controlled experiments further indicate that guidance relevance and path coherence contribute to these gains, highlighting the representation of privileged information as an important design choice in OPSD.

cs.AI↗

arXiv2026

Latent Thought Credit: Multi-Answer Credit Assignment for Latent Reasoning

Latent reasoning allows language models to carry out intermediate reasoning in continuous latent representations rather than fully externalizing it as discrete chains of thought. However, assigning credit to such latent thoughts from answer-only rewards is difficult: a single final answer mixes thought quality with answer-sampling noise. We propose \textbf{Latent Thought Credit (LTC)}, a hierarchical credit-assignment framework for latent reasoning. For each prompt, LTC samples multiple latent thoughts, fixes the context after each thought, and estimates thought-level expected reward by averaging rewards over multiple answers generated from that fixed context. LTC uses thought-level advantages to optimize the latent-thought phase, answer-level advantages to optimize the answer phase, and an advantage-weighted thought-matching objective that helps the policy reproduce high-credit latent thoughts. We instantiate LTC in a GRPO-style on-policy training framework and evaluate it across mathematical reasoning and STEM multiple-choice tasks. LTC achieves the best average accuracy among the compared methods, while ablations and fixed-context diagnostics show that multi-answer estimation reduces reward-estimation error and mitigates ambiguous or incorrect thought-level credit.

cs.AI↗

arXiv2026

TARPO: Token-Wise Latent-Explicit Reasoning via Action-Routing Policy Optimization

Latent reasoning has emerged as a promising alternative to discrete Chain-of-Thought (CoT) in large language models (LLMs), enabling more expressive reasoning by operating over continuous representations. However, the inherently deterministic nature of continuous representations limits policy exploration in reinforcement learning (RL). To address this, we propose TARPO (Token-Wise Latent-Explicit Reasoning via Action-Routing Policy Optimization), a pure RL framework that adaptively switches between discrete token generation and continuous latent reasoning at each step. TARPO introduces a lightweight action head router that observes the current hidden state and samples a routing decision from a binary mode-selection space, preserving the stochasticity of discrete token sampling from the vocabulary. The LLM backbone and router are jointly optimized end-to-end with a shared group-relative advantage signal. Extensive experiments across Qwen2.5 (from 1.5B to 7B) and Llama-3.1-8B backbones demonstrate that TARPO consistently outperforms existing explicit and latent reasoning RL baselines across diverse benchmarks. Further analysis shows that TARPO learns adaptive token-wise switching behaviors while maintaining stable training dynamics. Our code is available at https://github.com/NKU-LITI/TARPO-master.

cs.CL↗

arXiv2026

Large Language Model Post-Training: A Unified View of Off-Policy and On-Policy Learning

Post-training has become central to turning pretrained large language models (LLMs) into aligned, capable, and deployable systems. Recent progress spans supervised fine-tuning (SFT), preference optimization, reinforcement learning (RL), process supervision, verifier-guided methods, distillation, and multi-stage pipelines. Yet these methods are often discussed in fragmented ways, organized by labels or objectives rather than by the behavioral bottlenecks they address. This survey argues that LLM post-training is best understood as structured intervention on model behavior. We organize the field first by trajectory provenance, which defines two primary regimes: off-policy learning on externally supplied trajectories and on-policy learning on learner-generated rollouts. We then interpret methods through two recurring roles -- effective support expansion, which makes useful behaviors more reachable, and policy reshaping, which improves behavior within already reachable regions -- together with a complementary systems-level role, behavioral consolidation, which preserves, transfers, and amortizes useful behavior across stages and model transitions. Under this view, SFT may serve either support expansion or policy reshaping; preference optimization is usually off-policy reshaping, though online variants move closer to learner-generated states. On-policy RL often improves behavior on learner-generated states, but stronger guidance can also make hard-to-reach reasoning paths reachable. Distillation is often better understood as consolidation rather than only compression, and hybrid pipelines emerge as coordinated multi-stage compositions. Overall, the framework helps diagnose post-training bottlenecks and reason about stage composition, suggesting that progress increasingly depends on coordinated systems design rather than any single dominant objective.

cs.CL↗

arXiv2025

MAPEX: A Multi-Agent Pipeline for Keyphrase Extraction

Keyphrase extraction is a fundamental task in natural language processing. However, existing unsupervised prompt-based methods for Large Language Models (LLMs) often rely on single-stage inference pipelines with uniform prompting, regardless of document length or LLM backbone. Such one-size-fits-all designs hinder the full exploitation of LLMs' reasoning and generation capabilities, especially given the complexity of keyphrase extraction across diverse scenarios. To address these challenges, we propose MAPEX, the first framework that introduces multi-agent collaboration into keyphrase extraction. MAPEX coordinates LLM-based agents through modules for expert recruitment, candidate extraction, topic guidance, knowledge augmentation, and post-processing. A dual-path strategy dynamically adapts to document length: knowledge-driven extraction for short texts and topic-guided extraction for long texts. Extensive experiments on six benchmark datasets across three different LLMs demonstrate its strong generalization and universality, outperforming the state-of-the-art unsupervised method by 2.44% and standard LLM baselines by 4.01% in F1@5 on average. Code is available at https://github.com/NKU-LITI/MAPEX.

cs.CL↗

arXiv2025

AgentCDM: Enhancing Multi-Agent Collaborative Decision-Making via ACH-Inspired Structured Reasoning

Multi-agent systems (MAS) powered by large language models (LLMs) hold significant promise for solving complex decision-making tasks. However, the core process of collaborative decision-making (CDM) within these systems remains underexplored. Existing approaches often rely on either ``dictatorial" strategies that are vulnerable to the cognitive biases of a single agent, or ``voting-based" methods that fail to fully harness collective intelligence. To address these limitations, we propose \textbf{AgentCDM}, a structured framework for enhancing collaborative decision-making in LLM-based multi-agent systems. Drawing inspiration from the Analysis of Competing Hypotheses (ACH) in cognitive science, AgentCDM introduces a structured reasoning paradigm that systematically mitigates cognitive biases and shifts decision-making from passive answer selection to active hypothesis evaluation and construction. To internalize this reasoning process, we develop a two-stage training paradigm: the first stage uses explicit ACH-inspired scaffolding to guide the model through structured reasoning, while the second stage progressively removes this scaffolding to encourage autonomous generalization. Experiments on multiple benchmark datasets demonstrate that AgentCDM achieves state-of-the-art performance and exhibits strong generalization, validating its effectiveness in improving the quality and robustness of collaborative decisions in MAS.

cs.AI↗

arXiv2023

Deep learning radiomics for assessment of gastroesophageal varices in people with compensated advanced chronic liver disease

Objective: Bleeding from gastroesophageal varices (GEV) is a medical emergency associated with high mortality. We aim to construct an artificial intelligence-based model of two-dimensional shear wave elastography (2D-SWE) of the liver and spleen to precisely assess the risk of GEV and high-risk gastroesophageal varices (HRV). Design: A prospective multicenter study was conducted in patients with compensated advanced chronic liver disease. 305 patients were enrolled from 12 hospitals, and finally 265 patients were included, with 1136 liver stiffness measurement (LSM) images and 1042 spleen stiffness measurement (SSM) images generated by 2D-SWE. We leveraged deep learning methods to uncover associations between image features and patient risk, and thus conducted models to predict GEV and HRV. Results: A multi-modality Deep Learning Risk Prediction model (DLRP) was constructed to assess GEV and HRV, based on LSM and SSM images, and clinical information. Validation analysis revealed that the AUCs of DLRP were 0.91 for GEV (95% CI 0.90 to 0.93, p < 0.05) and 0.88 for HRV (95% CI 0.86 to 0.89, p < 0.01), which were significantly and robustly better than canonical risk indicators, including the value of LSM and SSM. Moreover, DLPR was better than the model using individual parameters, including LSM and SSM images. In HRV prediction, the 2D-SWE images of SSM outperform LSM (p < 0.01). Conclusion: DLRP shows excellent performance in predicting GEV and HRV over canonical risk indicators LSM and SSM. Additionally, the 2D-SWE images of SSM provided more information for better accuracy in predicting HRV than the LSM.

q-bio.TO↗