arXiv subjects

Pengwei Sun

Publications and source records attributed to Pengwei Sun.

7 recordsLinked to original sources

arXiv2026

Boosting Direct Preference Optimization with Penalization

Offline preference optimization has become a practical substitute for reinforcement learning from human feedback, but pairwise objectives such as Direct Preference Optimization (DPO) and its variants use only the chosen and rejected responses stored in a static dataset. This leaves a useful signal unused: the response that the reference model itself would generate for the same prompt. We propose Direct Preference Optimization with Penalization (DPOP), a simple extension of DPO that augments the base preference loss with a gated penalty on reference-greedy responses. DPOP activates this penalty only when the current policy still assigns a lower likelihood to the preferred response than to the rejected response. On AlpacaEval 2.0, DPOP improves length-controlled win rate over DPO, SimPO, and AlphaDPO on both Llama-3-8b-it and Gemma-2-9b-it, achieving relative gains of 5.3\% and 4.4\% over baselines on the two models, respectively. Ablations further show that a SimNPO-style length-normalized penalty is stronger than NPO and token-level unlikelihood in this setting.

cs.LG↗

arXiv2026

Structured Scaling of AI Discovery Across Diverse Scientific Domains

Scientific discovery often requires many cycles of proposing, testing, and refining candidate solutions. Language models can increasingly participate in these loops, but simply generating more attempts does not ensure progress: parallel searches may duplicate one another and iterative refinement may become trapped in poor directions. The central challenge is therefore not only to scale AI-driven discovery, but to structure that scaling so that evaluation signals compound over time. Here we introduce SimpleTES (Simple Test-time Evaluation-driven Scaling), a framework that focuses on the structured scaling of AI discovery loops, organizing evaluator queries across independent trajectories, iterative refinement, local candidate selection, and the selective reuse of evaluated histories. Drawing on structural features of scientific communities, SimpleTES uses a single open-source GPT-OSS model to establish new state-of-the-art solutions across 28 open-ended problems in diverse scientific domains ranging from quantum physics and astronomy to biology, AI, and mathematics. These include a 24.5% reduction in quantum circuit compilation overhead, up to 23% lower propulsive cost for deep-space trajectories, a 2.17x faster lasso-path solver, an 8.5% lower-error whole-brain neural-activity predictor, the fastest reported TriMul kernel, and new mathematical constructions beyond prior human or AI records. We further post-train the model for long-horizon discovery by assigning each attempt the final outcome of the trajectory it helped produce. This improves performance on both training and held-out mathematics problems, further advancing the frontier. Together, these results establish structured scaling as a general mechanism for advancing AI scientific discovery.

cs.LG↗

arXiv2026

Align Your Structures: Generating Trajectories with Structure Pretraining for Molecular Dynamics

Generating molecular dynamics (MD) trajectories using deep generative models has attracted increasing attention, yet remains inherently challenging due to the limited availability of MD data and the complexities involved in modeling high-dimensional MD distributions. To overcome these challenges, we propose a novel framework that leverages structure pretraining for MD trajectory generation. Specifically, we first train a diffusion-based structure generation model on a large-scale conformer dataset, on top of which we introduce an interpolator module trained on MD trajectory data, designed to enforce temporal consistency among generated structures. Our approach effectively harnesses abundant structural data to mitigate the scarcity of MD trajectory data and effectively decomposes the intricate MD modeling task into two manageable subproblems: structural generation and temporal alignment. We comprehensively evaluate our method on the QM9 and DRUGS small-molecule datasets across unconditional generation, forward simulation, and interpolation tasks, and further extend our framework and analysis to tetrapeptide and protein monomer systems. Experimental results confirm that our approach excels in generating chemically realistic MD trajectories, as evidenced by remarkable improvements of accuracy in geometric, dynamical, and energetic measurements.

cs.LG↗

arXiv2026

Curriculum Sampling: A Two-Phase Curriculum for Efficient Training of Flow Matching

Timestep sampling $p(t)$ is a central design choice in Flow Matching models, yet common practice increasingly favors static middle-biased distributions (e.g., Logit-Normal). We show that this choice induces a speed--quality trade-off: middle-biased sampling accelerates early convergence but yields worse asymptotic fidelity than Uniform sampling. By analyzing per-timestep training losses, we identify a U-shaped difficulty profile with persistent errors near the boundary regimes, implying that under-sampling the endpoints leaves fine details unresolved. Guided by this insight, we propose \textbf{Curriculum Sampling}, a two-phase schedule that begins with middle-biased sampling for rapid structure learning and then switches to Uniform sampling for boundary refinement. On CIFAR-10, Curriculum Sampling improves the best FID from $3.85$ (Uniform) to $3.22$ while reaching peak performance at $100$k rather than $150$k training steps. Our results highlight that timestep sampling should be treated as an evolving curriculum rather than a fixed hyperparameter.

cs.LG↗

arXiv2026

Microscopy of Bioelectric Potentials using Electrochromism

Studying the electrical signals generated by living cells is key to understanding numerous biological phenomena. Electrochromic optical recording (ECORE) uses the electrochromism exhibited by certain materials to noninvasively measure these signals in real time. In this work, we report on the development of ECORE based on a high-NA microscope objective. We demonstrate the recording of extracellular action potentials from cardiomyocytes with single-cell resolution and a high sensitivity of 6.3 {\mu}V. Combining ECORE with microscopy simplifies the optical setup, allows for the simultaneous imaging of specimens, and makes ECORE accessible to a broader community of researchers, allowing for a better understanding of the biological processes that are integral to life.

physics.optics↗

arXiv2025

Evaluation of 3D Counterfactual Brain MRI Generation

Counterfactual generation offers a principled framework for simulating hypothetical changes in medical imaging, with potential applications in understanding disease mechanisms and generating physiologically plausible data. However, generating realistic structural 3D brain MRIs that respect anatomical and causal constraints remains challenging due to data scarcity, structural complexity, and the lack of standardized evaluation protocols. In this work, we convert six generative models into 3D counterfactual approaches by incorporating an anatomy-guided framework based on a causal graph, in which regional brain volumes serve as direct conditioning inputs. Each model is evaluated with respect to composition, reversibility, realism, effectiveness and minimality on T1-weighted brain MRIs (T1w MRIs) from the Alzheimer's Disease Neuroimaging Initiative (ADNI). In addition, we test the generalizability of each model with respect to T1w MRIs of the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA). Our results indicate that anatomically grounded conditioning successfully modifies the targeted anatomical regions; however, it exhibits limitations in preserving non-targeted structures. Beyond laying the groundwork for more interpretable and clinically relevant generative modeling of brain MRIs, this benchmark highlights the need for novel architectures that more accurately capture anatomical interdependencies.

eess.IV↗

arXiv2022

Pseudo-Phase Transitions of Ising and Baxter-Wu Models in Two-Dimensional Finite-Size Lattices

This article offers a detailed analysis of pseudo-phase transitions of Ising and Baxter-Wu models in two-dimensional finite-size lattices. We carry out Wang Landau sampling to obtain the density of states. Using microcanonical inflection point analysis with microcanonical entropy, we obtain the order of the psuedo-phase transitions in the models. The microcanonical analysis results of the second-order transition for the Ising model and the first-order transition for the Baxter-Wu model are consistent with the traditional canonical results. In addition, the third-order transitions are found in both models, implying the universality of higher-order phase transitions.

cond-mat.stat-mech↗