arXiv subjects

Haipeng Chen

Publications and source records attributed to Haipeng Chen.

At least 19 recordsLinked to original sources

arXiv2026

DELA: Learning Controllable Edits for Diffusion Language Models

Diffusion language models (DLMs) support bidirectional denoising and iterative refinement, but generation remains largely controlled by token-level masking schedules. This leaves draft-level refinement implicit and limits explicit control over when to insert missing content, delete redundant spans, revisit uncertain regions, or stop editing. We introduce Diffusion Editing Language Adapter (DELA), a plug-in framework that enables controllable draft refinement with a frozen DLM. DELA operates on clean, variable-length drafts using four explicit editing actions: Insert, Delete, Remask, and Stop. The controller determines when and where to edit the draft, while the frozen DLM fills masked regions through denoising, separating high-level edit control from token-level generation. A naive single-phase editor would be computationally expensive because arbitrary interior edits can disrupt prefix stability and limit reliable KV-cache reuse. To address this, DELA uses a two-phase decoding procedure: a fast, append-only drafting phase that supports KV-cache reuse, followed by a global refinement phase that enables targeted non-monotonic edits to the full draft. We train the controller using a staged curriculum that begins with learning local edit priors from synthetic recovery tasks and mined actions that improve draft quality. Subsequent stages refine edit selection through rollout-based outcome preferences and alignment with downstream task feedback. Experiments on mathematical reasoning and code generation show that DELA achieves gains of up to 5.5 percentage points over direct diffusion decoding while keeping the backbone frozen. DELA thus provides a modular approach to explicit draft-level control in diffusion language models.

cs.AI↗

arXiv2026

AI-driven Prices for Externalities and Sustainability in Production Markets

Traditional competitive markets do not account for negative externalities; indirect costs that some participants impose on others, such as the cost of over-appropriating a common-pool resource (which diminishes future stock, and thus harvest, for everyone). Quantifying appropriate interventions to market prices has proven to be quite challenging. We propose a practical approach to computing market prices and allocations via a deep reinforcement learning policymaker agent, operating in an environment of other learning agents. Our policymaker allows us to tune the prices with regard to diverse objectives such as sustainability and resource wastefulness, fairness, buyers' and sellers' welfare, etc. As a highlight of our findings, our policymaker is significantly more successful in maintaining resource sustainability, compared to the market equilibrium outcome, in scarce resource environments.

cs.MA↗

arXiv2026

Denoising Blocks, Not Tokens: Efficient Compressed Continuous Diffusion with Branching Token Realization

Diffusion language models (DLMs) generate text through iterative parallel refinement, offering the potential for higher throughput than autoregressive (AR) decoding. However, most DLMs still maintain one generative state per token, so every denoising step processes a state sequence as long as the output sequence, limiting the throughput gains from parallel generation. Continuous DLMs provide an additional degree of freedom: a single continuous state can represent multiple tokens, allowing diffusion to operate on a much shorter latent sequence. We introduce \emph{Branching Latent Diffusion (BLD)}, which exploits this flexibility by compressing a 1024-token sequence into only 64 block latents, a $16\times$ reduction. BLD combines latent compression with \emph{branching token realization}, where each latent is decoded by a local AR branch and all branches run in parallel. Because strong compression makes joint latent generation difficult, BLD generates the latents in groups, conditioning each group on previously generated latents. In end-to-end evaluation on the same GPU, BLD reduces generation FLOPs by more than $80\times$ and increases throughput by more than $6\times$ relative to the similarly sized ELF-L baseline. Compared with the AR baseline, BLD achieves more than $6\times$ higher throughput and more than $4\times$ lower latency. Despite the compression, BLD maintains competitive local fluency and diversity, although long-range coherence remains challenging. Overall, BLD shows that moving diffusion from token-level states to compressed latent sequences can substantially improve the efficiency of long-sequence generation.

cs.LG↗

arXiv2026

Towards Reliable, Generalizable, and Specific In-Context Knowledge Editing via Multi-Objective Reinforcement Learning

Large Language Models (LLMs) are powerful but limited by static parametric knowledge that becomes outdated once pretraining ends. Knowledge editing addresses this problem by updating model behavior on target facts without full retraining. In particular, in-context knowledge editing has gained attention because it is training-free and readily applicable to black-box LLMs. Recent reinforcement learning (RL)-based approaches improve over fixed retrieval strategies by adapting prompt construction to the quantity-quality trade-off. Despite initial success, they fail to model the prompt as a structured entity under the distinct and often competing objectives of reliability, generality, and specificity. Previous methods largely optimize a single objective and make decisions over only part of the prompt construction process, thereby overlooking both the balance of different objectives and the global organization of demonstrations. We propose Multi-Objective In-context Knowledge Editing (MO-IKE), a multi-objective RL algorithm that formulates prompt construction for in-context knowledge editing as a Constrained Markov Decision Process. MO-IKE trains a dynamic retriever to optimize competing objectives in knowledge editing, enabling more balanced and globally coherent prompt construction. On Llama-3.2, MO-IKE improves edit success (reliability) from 85.0% to 92.0%, paraphrase consistency (generality) from 77% to 79%, while increasing retention rate (specificity) by 23.0% compared to prior RL-based methods.

cs.AI↗

arXiv2026

Exploring Forum Post Retrieval with Generative Modeling

Generative recommendation (GR) has emerged as an alternative to embedding-based retrieval, building on the success of generative models in language and vision. We are exploring GR on Facebook Forum, a standalone application for medium-to-heavy users of Facebook Groups. Because Forum is a new surface, its own interaction data are too sparse to train a GR model from scratch. We address this with transfer along two axes: we train on a broader corpus of Facebook Groups engagements rather than Forum sessions alone, and we reuse hierarchical, prefix-based semantic IDs (SIDs) learned from cross-platform Facebook Feed data instead of fitting a Forum-specific tokenizer. A 3B-parameter instruction-tuned language model is then supervised-fine-tuned to generate SIDs directly from user context. We systematically ablate the design choices that matter most in practice, including SID construction, the composition and length of user history, and the inclusion of user-profile features. Our results show that cross-platform SIDs transfer to a new recommendation surface, and offer practical guidance for teams deploying GR on real-world social platforms.

cs.IR↗

arXiv2026

VTV-FM: Flow Matching through Variational Terminal-Velocity Closure

Flow matching (FM) learns generative transport by fitting continuous-time motion from a simple source distribution to the data distribution. Most existing methods use first-order bridges: once a source and a target sample are paired, the path is a straight motion with constant velocity. FM with optimal transport (OT) improves the pairing, but the bridge itself remains linear, limiting its ability to model curved motion, acceleration, and changing directions. A natural remedy is to use second-order phase-space dynamics; however, learning the bridge requires target-side terminal-velocity information that static datasets do not provide. We propose Variational Terminal-Velocity Flow Matching (VTV-FM), a second-order FM framework that derives the missing velocity by minimizing acceleration energy, yielding a closed-form closure for static data. The same minimum-acceleration variational construction also defines the OT pairing cost and the acceleration targets used for training. Experiments on low-dimensional datasets, PDE-governed physical fields, and CIFAR-10 show that VTV-FM improves transport geometry and generation quality over first-order and high-order FM baselines.

cs.CV↗

arXiv2026

HyCO: A Hybrid Neural Solver for Combinatorial Optimization

Sequential reinforcement learning (RL) solvers and global diffusion model (DM) solvers for neural combinatorial optimization exhibit complementary failure modes under an optimization-regret view. The former enjoys small marginal regret in the early construction stage, but suffers from horizon-wise compounding errors with super-linear regret growth; the latter avoids horizon compounding but incurs linear or sublinear regret w.r.t. the dimension of the remaining unsolved subspace. We propose Hybrid Neural Solver for Combinatorial Optimization (HyCO), a hybrid inference algorithm that constructs a solution prefix with an RL solver and adaptively switches to a conditional DM to complete the remaining decisions. To characterize why such hybridization helps, when to trigger the handover, and how to realize it in practice, we first develop a unified error-scaling theoretical framework and prove that, under explicit error-scaling assumptions, i) the hybrid structure achieves strictly lower expected regret than either backbone alone, and ii) there exists a unique optimal trigger step that minimizes the hybrid regret. We then design a lightweight adaptive trigger that combines policy entropy and RL-DM disagreement to detect trajectory-level signals of the regime shift as a practical proxy, since the optimal trigger step is defined at the expected-regret level and is not directly computable on individual trajectories. Experimental results on diverse benchmarks demonstrate that HyCO achieves consistent improvements over both backbones and support the empirical effectiveness of adaptive triggering.

cs.LG↗

arXiv2026

DIP: Dynamic In-Context Planner For Diffusion Language Models

Diffusion language models (DLMs) have shown strong potential for general natural language tasks with in-context examples. Existing In-Context Learning (ICL) approaches largely inherit the practice of autoregressive language models (ARLMs), incorporating all examples into a fixed prompt. However, applying this rigid, static-prompt paradigm to DLMs incurs substantial computational overhead, as the model must evaluate the maximum context length at every step. We address this inefficiency with a key discovery: the block-wise KV-cache mechanism inherent to DLM inference enables the \textit{low-cost dynamic adjustment of the context}. Following this intuition, our core idea is to start generation with a minimal prompt and progressively insert additional examples on the fly only when the generated tokens are of low confidence. Through rigorous empirical evaluations, we observe that average verified token confidence correlates strongly with generation accuracy, making it a reliable and computationally efficient signal of token quality. Formally, we propose \textbf{D}ynamic \textbf{I}n-Context \textbf{P}lanner (DIP), a context-optimization algorithm based on average verified confidence that dynamically ranks and inserts in-context examples during generation, rather than providing all examples up front. Experimental results on math and coding benchmarks with LLaDA-1.5 and LLaDA-8B-Instruct show that DIP achieves up to $1.59\times$ and $1.36\times$ speedups, respectively, while largely preserving the generation quality of the fixed-prompt baseline. Code: https://github.com/wmd3i/DIP

cs.CL↗

arXiv2026

Offline Reinforcement Learning with Generative Trajectory Policies

Generative models have emerged as a powerful class of policies for offline reinforcement learning (RL) due to their ability to capture complex, multi-modal behaviors. However, existing methods face a stark trade-off: slow, iterative models like diffusion policies are computationally expensive, while fast, single-step models like consistency policies often suffer from degraded performance. In this paper, we demonstrate that it is possible to bridge this gap. The key to moving beyond the limitations of individual methods, we argue, lies in a unifying perspective that views modern generative models, including diffusion, flow matching, and consistency models, as specific instances of learning a continuous-time generative trajectory governed by an Ordinary Differential Equation (ODE). This principled foundation provides a clearer design space for generative policies in RL and allows us to propose Generative Trajectory Policies (GTPs), a new and more general policy paradigm that learns the entire solution map of the underlying ODE. To make this paradigm practical for offline RL, we further introduce two key theoretically principled adaptations. Empirical results demonstrate that GTP achieves state-of-the-art performance on D4RL benchmarks - it significantly outperforms prior generative policies, achieving perfect scores on several notoriously hard AntMaze tasks.

cs.LG↗

arXiv2026

Using Synthetic Data for Machine Learning-based Childhood Vaccination Prediction in Narok, Kenya

Background: Limited data utilization in low-resource settings poses a barrier to the vaccine delivery ecosystem, undermining efforts to achieve equitable immunization coverage. In nomadic populations, individuals face an increased risk of missing crucial vaccination doses as children. One such population is the Maasai in Narok County, Kenya, where the absence of high-volume, quality data hampers accurate coverage estimates, impedes efficient resource allocation, and weakens the ability to deliver timely interventions. Additionally, data privacy concerns are heightened in groups with limited sensitive data. Objectives: First, we aim to identify children at risk of missing key vaccines across a large population to provide timely, evidence-based interventions that support increased vaccination coverage. Second, we aim to better protect the privacy of sensitive health data in a vulnerable population. Methods: We digitized 8 years of child vaccination records from the MOH 510 registry (n=6,913) and applied machine learning models (Logistic Regression and XGBoost) to identify children at risk. Additionally, we utilize a novel approach to tabular diffusion-based synthetic data generation (TabSyn) to protect patient privacy within the models. Results: Our findings show that classification techniques can reliably and successfully predict children at risk of missing a vaccine, with recall, precision, and F1-scores exceeding 90% for some vaccines modeled. Additionally, training these models with synthetic data rather than real data, thus preserving the privacy of individuals within the original dataset, does not lead to a loss in predictive performance. Conclusion: These results support the use of synthetic data implementation in health informatics strategies for clinics with limited digital infrastructure, enabling privacy-preserving, scalable forecasting for childhood immunization coverage.

cs.LG↗

arXiv2026

Attainable forms of lower spectra

Let $d\in\mathbb{N}$ and $φ\colon(0,1)\to[0,d]$. We prove there exists a set $F\subset\mathbb{R}^d$ whose lower spectrum $\operatorname{dim}^θ_{\mathrm{L}} F$ satisfies $(1-θ)\operatorname{dim}^θ_{\mathrm{L}} F = φ(θ)$ for all $θ\in(0,1)$ if and only if for all $λ,θ\in(0,1)$, \begin{equation*} φ(θ) \leq φ(λθ) - θφ(λ) \leq (1-θ) d. \end{equation*} We also obtain a similar classification result for $\underline{\operatorname{dim}}^θ_{\mathrm{L}} F$. In contrast to the case for Assouad spectra, it is insufficient to consider homogeneous (or uniform) sets. Instead, we follow the approach introduced by Orgoványi--Rutar in arXiv:2510.07013 and proceed via a more general classification result for appropriate two-scale branching functions.

math.CA↗

arXiv2026

Generalised lower Assouad-type dimensions and their interpolations

This paper investigates the analytic and structural properties of the $ϕ$-lower Assouad dimension, a generalized notion extending the lower Assouad dimension. We establish the equivalence of $ϕ$-lower Assouad dimensions with respect to the dimension functions, prove analytic properties related to the regularity of the $ϕ$-lower dimension, and analyse the role of rate windows in this context. Furthermore, we explore both positive and negative interpolation properties of the $ϕ$-lower dimension by presenting corresponding theorems that delineate these behaviors.

math.CA↗

arXiv2026

Speculative Sampling with Reinforcement Learning

Inference time latency has remained an open challenge for real world applications of large language models (LLMs). State-of-the-art (SOTA) speculative sampling (SpS) methods for LLMs, like EAGLE-3, use tree-based drafting to explore multiple candidate continuations in parallel. However, the hyperparameters controlling the tree structure are static, which limits flexibility and efficiency across diverse contexts and domains. We introduce Reinforcement learning for Speculative Sampling (Re-SpS), the first reinforcement learning (RL)-based framework for draft tree hyperparameter optimization. Re-SpS dynamically adjusts draft tree hyperparameters in real-time, learning context-aware policies that maximize generation speed by balancing speculative aggression with computational overhead. It leverages efficient state representations from target model hidden states and introduces multi-step action persistence for better context modeling. Evaluation results across five diverse benchmarks demonstrate consistent improvements over the SOTA method EAGLE-3, achieving up to 5.45$\times$ speedup over the backbone LLM and up to 1.12$\times$ speedup compared to EAGLE-3 across five diverse benchmarks, with no loss in output fidelity.

cs.LG↗

arXiv2025

Representations of rational numbers and Minkowski dimension

In this paper, we investigate the representations of rational numbers via continued fraction, Egyptian fraction, and Engel fraction expansions. Given $m \in \mathbb{N}$, denote by $C_m, E_m, E_m^*$ the sets of rational numbers whose continued fraction, Egyptian fraction, and Engel fraction expansions have length $m$, respectively. We first establish the Minkowski dimensions of these sets, which implies that their global scaling properties are different. We also apply the results to sumsets of decreasing sequences.

math.CA↗

arXiv2025

Dynamic Retriever for In-Context Knowledge Editing via Policy Optimization

Large language models (LLMs) excel at factual recall yet still propagate stale or incorrect knowledge. In-context knowledge editing offers a gradient-free remedy suitable for black-box APIs, but current editors rely on static demonstration sets chosen by surface-level similarity, leading to two persistent obstacles: (i) a quantity-quality trade-off, and (ii) lack of adaptivity to task difficulty. We address these issues by dynamically selecting supporting demonstrations according to their utility for the edit. We propose Dynamic Retriever for In-Context Knowledge Editing (DR-IKE), a lightweight framework that (1) trains a BERT retriever with REINFORCE to rank demonstrations by editing reward, and (2) employs a learnable threshold to prune low-value examples, shortening the prompt when the edit is easy and expanding it when the task is hard. DR-IKE performs editing without modifying model weights, relying solely on forward passes for compatibility with black-box LLMs. On the COUNTERFACT benchmark, it improves edit success by up to 17.1%, reduces latency by 41.6%, and preserves accuracy on unrelated queries, demonstrating scalable and adaptive knowledge editing. The code is available at https://github.com/mwnafee/DR-IKE .

cs.CL↗

arXiv2025

Recursive Gaussian Process State Space Model

Learning dynamical models from data is not only fundamental but also holds great promise for advancing principle discovery, time-series prediction, and controller design. Among various approaches, Gaussian Process State-Space Models (GPSSMs) have recently gained significant attention due to their combination of flexibility and interpretability. However, for online learning, the field lacks an efficient method suitable for scenarios where prior information regarding data distribution and model function is limited. To address this issue, this paper proposes a recursive GPSSM method with adaptive capabilities for both operating domains and Gaussian process (GP) hyperparameters. Specifically, we first utilize first-order linearization to derive a Bayesian update equation for the joint distribution between the system state and the GP model, enabling closed-form and domain-independent learning. Second, an online selection algorithm for inducing points is developed based on informative criteria to achieve lightweight learning. Third, to support online hyperparameter optimization, we recover historical measurement information from the current filtering distribution. Comprehensive evaluations on both synthetic and real-world datasets demonstrate the superior accuracy, computational efficiency, and adaptability of our method compared to state-of-the-art online GPSSM techniques.

cs.LG↗

arXiv2025

Dual-supervised Asymmetric Co-training for Semi-supervised Medical Domain Generalization

Semi-supervised domain generalization (SSDG) in medical image segmentation offers a promising solution for generalizing to unseen domains during testing, addressing domain shift challenges and minimizing annotation costs. However, conventional SSDG methods assume labeled and unlabeled data are available for each source domain in the training set, a condition that is not always met in practice. The coexistence of limited annotation and domain shift in the training set is a prevalent issue. Thus, this paper explores a more practical and challenging scenario, cross-domain semi-supervised domain generalization (CD-SSDG), where domain shifts occur between labeled and unlabeled training data, in addition to shifts between training and testing sets. Existing SSDG methods exhibit sub-optimal performance under such domain shifts because of inaccurate pseudolabels. To address this issue, we propose a novel dual-supervised asymmetric co-training (DAC) framework tailored for CD-SSDG. Building upon the co-training paradigm with two sub-models offering cross pseudo supervision, our DAC framework integrates extra feature-level supervision and asymmetric auxiliary tasks for each sub-model. This feature-level supervision serves to address inaccurate pseudo supervision caused by domain shifts between labeled and unlabeled data, utilizing complementary supervision from the rich feature space. Additionally, two distinct auxiliary self-supervised tasks are integrated into each sub-model to enhance domain-invariant discriminative feature learning and prevent model collapse. Extensive experiments on real-world medical image segmentation datasets, i.e., Fundus, Polyp, and SCGM, demonstrate the robust generalizability of the proposed DAC framework.

cs.CV↗

arXiv2025

HVIS: A Human-like Vision and Inference System for Human Motion Prediction

Grasping the intricacies of human motion, which involve perceiving spatio-temporal dependence and multi-scale effects, is essential for predicting human motion. While humans inherently possess the requisite skills to navigate this issue, it proves to be markedly more challenging for machines to emulate. To bridge the gap, we propose the Human-like Vision and Inference System (HVIS) for human motion prediction, which is designed to emulate human observation and forecast future movements. HVIS comprises two components: the human-like vision encode (HVE) module and the human-like motion inference (HMI) module. The HVE module mimics and refines the human visual process, incorporating a retina-analog component that captures spatiotemporal information separately to avoid unnecessary crosstalk. Additionally, a visual cortex-analogy component is designed to hierarchically extract and treat complex motion features, focusing on both global and local features of human poses. The HMI is employed to simulate the multi-stage learning model of the human brain. The spontaneous learning network simulates the neuronal fracture generation process for the adversarial generation of future motions. Subsequently, the deliberate learning network is optimized for hard-to-train joints to prevent misleading learning. Experimental results demonstrate that our method achieves new state-of-the-art performance, significantly outperforming existing methods by 19.8% on Human3.6M, 15.7% on CMU Mocap, and 11.1% on G3D.

cs.CV↗