arXiv subjects

Chenliang Li

Publications and source records attributed to Chenliang Li.

At least 19 recordsLinked to original sources

arXiv2026

Building Presentation Intelligence for High-Quality Slide Generation

PowerPoint presentations (PPTs) are a dominant medium for communication across diverse domains, yet high-quality PPT generation still depends heavily on labor-intensive human effort. Recent advances in large language models (LLMs) for agentic tasks have created both new opportunities and challenges (Chen et al., 2026) for automatic PPT generation. In this work, we present a unified training framework for automatic PPT generation toward presentation intelligence, integrating cross-validated data synthesis, guided teacher enhancement, and verifiable reward optimization. We first construct a cross-validated pipeline for data synthesis, with explicit control over both difficulty and validity. We then develop a teacher-enhancement framework grounded in rules and prior knowledge to substantially strengthen the instructional capacity of teacher models. Building on this, we further improve supervised fine-tuning (SFT) through a mixed-teacher strategy and a lightweight outcome-driven trajectory filtering method. Finally, based on the resulting SFT models, we introduce a suite of verifiable aesthetic metrics together with a GRPO-based reinforcement learning framework to optimize slide-level visual attributes beyond text-only supervision. Experiments show that our enhanced SFT pipeline improves Qwen-3.6-27B and Qwen-3.6-35B-A3B by 7.45 and 9.35 points on PresentBench, respectively, surpassing GLM-5.2 while maintaining strong generalization. Reinforcement learning further improves performance across five aesthetic dimensions, compensating for the limitations of SFT in visual refinement. Overall, the results demonstrate the effectiveness of our framework for high-quality PPT generation.

cs.LG↗

arXiv2026

Aligning Large Language Models with Human Feedback: Mathematical Foundations and Algorithm Design

This article provides an introduction to the mathematical foundations and algorithmic frameworks used to align Large Language Models (LLMs) with human intentions, preferences, and values. We discuss standard alignment techniques, such as fine-tuning (SFT), reinforcement learning with human feedback (RLHF), and direct preference optimization (DPO). We also explore the theoretical underpinnings of learning from human preferences, drawing connections to inverse reinforcement learning (IRL) and discrete choice models. We present state-of-the-art algorithms in a tutorial style, discuss their advantages and limitations, and offer insights into practical implementation. Our exposition is intended to serve as a comprehensive resource for researchers and practitioners, providing both a foundational understanding of alignment methodologies and a framework for developing more robust and scalable alignment techniques.

cs.LG↗

arXiv2026

Stabilizing Off-Policy Training for Long-Horizon LLM Agent via Turn-Level Importance Sampling and Clipping-Triggered Normalization

Reinforcement learning (RL) algorithms such as PPO and GRPO are widely used to train large language models (LLMs) for multi-turn agentic tasks. However, in off-policy training pipelines, these methods can exhibit unstable optimization dynamics and are prone to perfor- mance collapse. Through empirical analysis, we identify two fundamental sources of instability in this setting: (1) a granularity mismatch between token-level policy optimization and turn- structured interactions, and (2) high-variance and unreliable gradient updates induced by off- policy importance sampling and inaccurate advantage estimation. To address these challenges, we propose SORL, Stabilizing Off-Policy Reinforcement Learning for Long-Horizon Agent Train- ing. SORL introduces mechanisms that align policy optimization with the structure of multi- turn interactions and adaptively suppress unreliable off-policy updates, yielding more conserva- tive and robust learning dynamics. Within this framework, we instantiate two stabilized algo- rithms: SO-PPO and SO-GRPO. Both algorithms are designed to mitigate gradient variance and prevent optimization collapse without requiring careful early stopping or heuristic tuning. We evaluate SO-PPO and SO-GRPO on benchmarks spanning open-domain QA, multi-hop QA, and medical multiple-choice QA, and further assess their transfer to asynchronous RL for mathe- matical reasoning by training on DAPO-Math-17k and validating on AIME-2024. These results demonstrate that SORL provides a practical, scalable, and general framework for stabilizing re- inforcement learning in multi-turn LLM agent training and asynchronous RL for mathematical reasoning.

cs.LG↗

arXiv2026

Where Does Staleness Accumulate? Pool Aware Effective Staleness Control for Asynchronous RL in LLM Post-Training

Fully asynchronous reinforcement learning (RL) improves resource utilization in large language model post-training by overlapping rollout generation with policy optimization, but it also introduces policy lag as trajectories are generated and queued while the trainer continues to update. We study how this lag accumulates over a trajectory's lifetime and how it can be controlled without sacrificing the wall-clock benefits of asynchronous execution. We decompose trajectory staleness into Generation Staleness, accumulated before rollout completion, and Waiting Staleness, accumulated after a completed trajectory enters the pool. Motivated by this decomposition, we introduce PACE (Pool-Aware Control of Effective Staleness). PACE converts excess pool occupancy into an adaptive rejection budget and ranks completed trajectories using an effective-staleness score that combines Waiting Staleness with prefix-aware Generation Staleness. This avoids penalizing long or interrupted rollouts solely because they span multiple policy versions. In single-turn mathematical reasoning, PACE improves the six-benchmark average validation accuracy by 18.7\% over unfiltered asynchronous RL at the same wall-clock budget and matches synchronous RL performance with 47.1\% less GPU time. PACE also improves validation performance in multi-turn tool-integrated reasoning, outperforming both synchronous and unfiltered asynchronous RL. Further experiments with the mixture-of-experts model and an alternative RL algorithm support its applicability across model architectures and training algorithms.

cs.AI↗

arXiv2026

GSAR: Goal-State-Anchor Rewards for Mobile GUI Agents with Self-Evolving Data Synthesis

Vision-Language Models (VLMs) based GUI agents stand to benefit significantly from online reinforcement learning (RL). However, their training is bottlenecked by two fundamental issues: current data synthesis methods for GUI Agents rely on specific environments and struggle to generate diverse data, while existing evaluators either suffer from limited scalability or provide inaccurate and unreliable reward signals. To overcome these challenges, we introduce GSAR (Goal-State-Anchor Reward), a RL reward framework that supports scalable task generation and delivers reliable reward signals for stable and efficient policy optimization. Our approach features self-evolving data synthesis, which produces multiple environments through task execution and generates diverse tasks and goal states. Complementing this, a state-anchor mechanism automatically annotates task-relevant UI elements in successful goal states as reference anchors. During RL training, these reference anchors provide accurate, scalable reward signals that substantially enhance efficiency. Extensive evaluations demonstrate that our framework achieves over 90% accuracy on offline trajectory verification and performs closest to rule-based methods. Furthermore, agents trained using our reward framework exhibit strong performance on both AndroidWorld and our constructed benchmark, establishing a scalable approach for GUI agent training.

cs.AI↗

arXiv2026

Reinforcing Multi-Turn Reasoning in LLM Agents via Fine-Grained Reward Structure and Credit Assignment

Reinforcement Learning (RL) approaches have been wildly used to enhance the reasoning capabilities of Large Language Model (LLM) agents in long-horizon, multi-turn scenarios. Such interactions can be formalized as turn-level Markov decision processes (MDPs), where intermediate rewards are often available. However, most prior work relies on sparse trajectory-level rewards, resulting in poor credit assignment, while dense turn-level rewards remain underexplored. In this paper, we investigate how to effectively leverage dense turn-level reward structures in RL algorithms, specifically Group Relative Policy Optimization (GRPO) and Proximal Policy Optimization (PPO), to enable fine-grained credit assignment. We categorize reward structures into three types based on their granularity: (1) terminal reward; (2) delayed reward; (3) per-turn reward, each corresponding to a distinct turn-level MDP formulation, and derive GRPO and PPO algorithms tailored to each case, respectively. Experiments on multi-turn search and game agent tasks show that, for both GRPO and PPO, dense per-turn reward structures consistently outperform sparse terminal and delayed reward structures in terms of training dynamics and numerical results. Furthermore, on search tasks, PPO with dense per-turn rewards achieves greater training stability and faster convergence, and delivers the highest answer correctness among all state-of-the-art methods across diverse question-answering datasets.

cs.LG↗

arXiv2026

Benchmarking General Mobile Assistants in Challenging Real-World Scenarios

Graphical user interfaces have emerged as an important environment for evaluating autonomous AI agents on multimodal interactive tasks. Existing benchmarks such as AndroidWorld and MobileWorld provide strong foundations for mobile agent evaluation, but their application coverage and task design do not yet fully capture the diversity and complexity of realistic mobile use. We present GMA, a benchmark for evaluating general mobile assistants in challenging real-world scenarios. GMA introduces seven applications based on open-source projects, spanning domains such as lifestyle sharing and travel planning, and 300 tasks across four difficulty tiers, from atomic actions to complex multi-step workflows. We evaluate eight frontier models and find that performance declines substantially as task complexity increases, with current agents remaining far from reliably handling realistic user requirements. We further conduct controlled ablation studies of agent harness choices, including context retention and explicit state tracking, under a shared environment, model setting, and task taxonomy. Results show that appropriate harness design can meaningfully improve performance, particularly on demanding workflows, while the effectiveness of specific designs can vary across foundation models. Overall, GMA complements existing benchmarks by expanding application coverage and task complexity, providing a challenging testbed for evaluating mobile agents and studying how harness design supports reliable execution in complex mobile workflows.

cs.AI↗

arXiv2026

Q-Regularized Generative Auto-Bidding: From Suboptimal Trajectories to Optimal Policies

With the rapid development of e-commerce, auto-bidding has become a key asset in optimizing advertising performance under diverse advertiser environments. The current approaches focus on reinforcement learning (RL) and generative models. These efforts imitate offline historical behaviors by utilizing a complex structure with expensive hyperparameter tuning. The suboptimal trajectories further exacerbate the difficulty of policy learning. To address these challenges, we proposes QGA, a novel Q-value regularized Generative Auto-bidding method. In QGA, we propose to plug a Q-value regularization with double Q-learning strategy into the Decision Transformer backbone. This design enables joint optimization of policy imitation and action-value maximization, allowing the learned bidding policy to both leverage experience from the dataset and alleviate the adverse impact of the suboptimal trajectories. Furthermore, to safely explore the policy space beyond the data distribution, we propose a Q-value guided dual-exploration mechanism, in which the DT model is conditioned on multiple return-to-go targets and locally perturbed actions. This entire exploration process is dynamically guided by the aforementioned Q-value module, which provides principled evaluation for each candidate action. Experiments on public benchmarks and simulation environments demonstrate that QGA consistently achieves superior or highly competitive results compared to existing alternatives. Notably, in large-scale real-world A/B testing, QGA achieves a 3.27% increase in Ad GMV and a 2.49% improvement in Ad ROI.

cs.LG↗

arXiv2026

State2State: Environment-Derived Mid-Training for LLM Agents

Training LLM agents commonly relies on supervised fine-tuning from expert trajectories or online reinforcement learning over human-specified tasks with handcrafted verifiers. Though effective, both remain bottlenecked by externally specified tasks and supervision signals, limiting the scalability and diversity of agent training. We study an environment learning paradigm in which agents acquire interaction and manipulation capabilities solely through environment interaction, without externally specified tasks. We propose State2State, an environment-derived mid-training method that converts explored environment states into training objectives, challenging agents to reach a specified target state. By deriving tasks from environment exploration and verifying success through rule-based state matching, State2State provides scalable and verifiable training objectives without expert supervision or manual task design. Experiments on ALFWorld and ScienceWorld show that State2State improves agent performance as a standalone environment-learning stage in most settings. As initialization for downstream RL, it further improves final performance and learning efficiency, with promising evidence of cross-environment generalization.

cs.CL↗

arXiv2026

An economic cascadic tensor multigrid method for solving high dimensional elliptic linear partial differential problems

In this paper, based on the tensor form, we propose a class of economic cascadic tensor multigrid method(ECTMG) for solving high dimensional elliptic linear partial differential problems. Compared with traditional methods, the new method not only reduces the storage space but also lowers the computational complexity from $\mathcal{O}(n^{3})$ to $\mathcal{O}(n^{2})$. We analyze the convergence rate of the conjugate gradient method which is based on the tensor form($\mathrm{CG}\_\mathrm{BTF}$), and then provide the convergence analysis for the new method. Finally, the effectiveness of the new method is verified through numerical examples.

math.NA↗

arXiv2026

Generative Auto-Bidding with Unified Modeling and Exploration

Automated bidding is central to modern digital advertising. Early rule-based methods lacked adaptability, while subsequent Reinforcement Learning approaches modeled bidding as a Markov Decision Process but struggled with long-term dependencies. Recent generative models show promise, yet they lack explicit mechanisms to balance exploration and safety, relying solely on action perturbations or trajectory guidance without a safety fallback. This results in inefficient exploration and elevated financial risk for advertising platforms. To address this gap, we propose GUIDE (Generative Auto-Bidding with Unified Modeling and Exploration), a framework that synergistically integrates directed exploration with a safe fallback mechanism. GUIDE employs a Decision Transformer (DT) to jointly model historical bidding actions and environmental state transitions. A Q-value module guides the DT's exploration via regularization constraints, while an Inverse Dynamics Module (IDM) leverages DT-predicted future states to infer robust, behaviorally consistent actions as a safe policy fallback. The Q-value module then adaptively selects the final action between these two options, balancing exploration and safety. Together, these components form an integrated "explore-safeguard-select" pipeline that unifies efficiency and safety. We conduct extensive experiments on public datasets, in simulated auction environments, and through large-scale online deployment on Taobao, a leading Chinese advertising platform. Results show GUIDE consistently outperforms state-of-the-art baselines across all scenarios. In real-world deployment, GUIDE achieves notable gains: +4.10% ad GMV, +1.40% ad clicks, +1.66% ad cost, and +3.52% ad ROI, demonstrating its effectiveness and strong industrial applicability.

cs.AI↗

arXiv2026

Origin-Destination Demand Prediction: An Urban Radiation and Attraction Perspective

In recent years, origin-destination (OD) demand prediction has gained significant attention for its profound implications in urban development. Existing data-driven deep learning methods primarily focus on the spatial or temporal dependency between regions yet neglecting regions' fundamental functional difference. Though knowledge-driven physical methods have characterised regions' functions by their radiation and attraction capacities, these functions are defined on numerical factors like population without considering regions' intrinsic nominal attributes, e.g., a region is a residential or industrial district. Moreover, the complicated relationships between two types of capacities, e.g., the radiation capacity of a residential district in the morning will be transformed into the attraction capacity in the evening, are totally missing from physical methods. In this paper, we not only generalize the physical radiation and attraction capacities into the deep learning framework with the extended capability to fulfil regions' functions, but also present a new model that captures the relationships between two types of capacities. Specifically, we first model regions' radiation and attraction capacities using a bilateral branch network, each equipped with regions' attribute representations. We then describe the transformation relationship of different capacities of the same region using a hypergraph-based parameter generation method. We finally unveil the competition relationship of different regions with the same attraction capacity through cluster-based adversarial learning. Extensive experiments on two datasets demonstrate the consistent improvements of our method over the state-of-the-art baselines, as well as the good explainability of regions' functions using their nominal attributes.

cs.LG↗

arXiv2026

CorpusQA: A 10 Million Token Benchmark for Corpus-Level Analysis and Reasoning

While large language models now handle million-token contexts, their capacity for reasoning across entire document repositories remains largely untested. Existing benchmarks are inadequate, as they are mostly limited to single long texts or rely on a "sparse retrieval" assumption-that answers can be derived from a few relevant chunks. This assumption fails for true corpus-level analysis, where evidence is highly dispersed across hundreds of documents and answers require global integration, comparison, and statistical aggregation. To address this critical gap, we introduce CorpusQA, a new benchmark scaling up to 10 million tokens, generated via a novel data synthesis framework. By decoupling reasoning from textual representation, this framework creates complex, computation-intensive queries with programmatically guaranteed ground-truth answers, challenging systems to perform holistic reasoning over vast, unstructured text without relying on fallible human annotation. We further demonstrate the utility of our framework beyond evaluation, showing that fine-tuning on our synthesized data effectively enhances an LLM's general long-context reasoning capabilities. Extensive experiments reveal that even state-of-the-art long-context LLMs struggle as input length increases, and standard retrieval-augmented generation systems collapse entirely. Our findings indicate that memory-augmented agentic architectures offer a more robust alternative, suggesting a critical shift is needed from simply extending context windows to developing advanced architectures for global information synthesis.

cs.CL↗

arXiv2026

Writing-RL: Advancing Long-form Writing via Adaptive Curriculum Reinforcement Learning

Recent advances in Large Language Models(LLMs) have enabled strong performance in long-form writing, but current training paradigms remain limited: Supervised Fine-Tuning (SFT) remains constrained by data saturation and performance ceilings, while Reinforcement Learning with Verifiable Reward (RLVR), though successful in verifiable domains like math and code, cannot be directly migrated to open-ended long-form writing due to a lack of ground-truths. To further advance long-form writing, we present Writing-RL: an Adaptive Curriculum Reinforcement Learning framework to advance long-form writing capabilities beyond SFT. The framework consists of three key components: Margin-aware Data Selection strategy that prioritizes samples with high learning potential, Pairwise Comparison Reward mechanism that provides discriminative learning signals in the absence of verifiable rewards, and Dynamic Reference Scheduling approach, which plays a critical role by adaptively adjusting task difficulty based on evolving model performance. Experiments on 7B-scale writer models show that Writing-RL effectively improves long-form writing performance over strong SFT baselines. Furthermore, we observe that models trained with long-output RL generalize surprisingly well to long-input reasoning tasks, potentially offering a promising perspective for rethinking long-context training.

cs.CL↗

arXiv2026

MUSEG: Reinforcing Video Temporal Understanding via Timestamp-Aware Multi-Segment Grounding

Video temporal understanding is crucial for multimodal large language models (MLLMs) to reason over events in videos. Despite recent advances in general video understanding, current MLLMs still struggle with fine-grained temporal reasoning. While reinforcement learning (RL) has been explored to address this issue recently, existing RL approaches remain limited in performance on time-sensitive tasks. In this work, we propose MUSEG, a novel RL-based method that enhances temporal understanding by introducing timestamp-aware multi-segment grounding. MUSEG enables MLLMs to align queries with multiple relevant video segments, promoting more comprehensive temporal reasoning. To facilitate effective learning, we design a customized RL training recipe with phased rewards that progressively guides the model toward temporally grounded reasoning. Extensive experiments on temporal grounding and time-sensitive video question answering (QA) tasks demonstrate that MUSEG significantly outperforms existing methods and generalizes well across diverse temporal understanding scenarios. View our project at https://github.com/THUNLP-MT/MUSEG.

cs.CV↗

arXiv2026

R2-Write: Reflection and Revision for Open-Ended Writing with Deep Reasoning

While deep reasoning with long chain-of-thought has dramatically improved large language models in verifiable domains like mathematics, its effectiveness for open-ended tasks such as writing remains unexplored. In this paper, we conduct a systematic investigation revealing that existing mainstream reasoning models achieve limited gains on open-ended writing tasks. Our further analysis shows that these models lack deep reflection and revision patterns in open-ended writing, resulting in substantially smaller improvements compared to mathematical reasoning tasks. To address this limitation, we introduce R2-Write: an automated framework that synthesizes high-quality thinking trajectories enriched with explicit reflection and revision patterns through iterative writer-judge interaction. To prevent redundant reflections, we design a process reward mechanism that supervises reflection quality during reinforcement learning, improving both performance and token efficiency. Extensive experiments across multiple creative writing and deep-research benchmarks demonstrate significant improvements, validating that explicitly incorporating reflection and revision patterns unlocks deep reasoning capabilities for open-ended writing tasks.

cs.CL↗

arXiv2026

Enhancing Pretrained Model-based Continual Representation Learning via Guided Random Projection

Recent paradigms in Random Projection Layer (RPL)-based continual representation learning have demonstrated superior performance when building upon a pre-trained model (PTM). These methods insert a randomly initialized RPL after a PTM to enhance feature representation in the initial stage. Subsequently, a linear classification head is used for analytic updates in the continual learning stage. However, under severe domain gaps between pre-trained representations and target domains, a randomly initialized RPL exhibits limited expressivity under large domain shifts. While largely scaling up the RPL dimension can improve expressivity, it also induces an ill-conditioned feature matrix, thereby destabilizing the recursive analytic updates of the linear head. To this end, we propose the Stochastic Continual Learner with MemoryGuard Supervisory Mechanism (SCL-MGSM). Unlike random initialization, MGSM constructs the projection layer via a principled, data-guided mechanism that progressively selects target-aligned random bases to adapt the PTM representation to downstream tasks. This facilitates the construction of a compact yet expressive RPL while improving the numerical stability of analytic updates. Extensive experiments on multiple exemplar-free Class Incremental Learning (CIL) benchmarks demonstrate that SCL-MGSM achieves superior performance compared to state-of-the-art methods.

cs.LG↗

arXiv2026

SPELL: Self-Play Reinforcement Learning for Evolving Long-Context Language Models

Progress in long-context reasoning for large language models (LLMs) has lagged behind other recent advances. This gap arises not only from the intrinsic difficulty of processing long texts, but also from the scarcity of reliable human annotations and programmatically verifiable reward signals. In this paper, we propose SPELL, a multi-role self-play reinforcement learning framework that enables scalable, label-free optimization for long-context reasoning. SPELL integrates three cyclical roles-questioner, responder, and verifier-within a single model to enable continual self-improvement. The questioner generates questions from raw documents paired with reference answers; the responder learns to solve these questions based on the documents; and the verifier evaluates semantic equivalence between the responder's output and the questioner's reference answer, producing reward signals to guide continual training. To stabilize training, we introduce an automated curriculum that gradually increases document length and a reward function that adapts question difficulty to the model's evolving capabilities. Extensive experiments on six long-context benchmarks show that SPELL consistently improves performance across diverse LLMs and outperforms equally sized models fine-tuned on large-scale annotated data. Notably, SPELL achieves an average 7.6-point gain in pass@8 on the strong reasoning model Qwen3-30B-A3B-Thinking, raising its performance ceiling and showing promise for scaling to even more capable models. Our code is available at https://github.com/Tongyi-Zhiwen/Qwen-Doc.

cs.CL↗