One of the biggest missing capabilities in current AI systems is the ability to learn continuously after deployment. Implementing such continually learning systems have several challenges, one of which is the large memory requirement of gradient-based algorithms that are used to train state-of-the-art LLMs. Evolutionary Strategies (ES) have recently re-emerged as a gradient-free alternative to traditional learning algorithms and have shown encouraging performance on specific tasks in LLMs. In this paper, we perform a comprehensive analysis of ES and specifically evaluate its forgetting curves when training for an increasing number of update steps. We first find that ES is able to reach performance numbers close to GRPO for math and reasoning tasks with a comparable compute budget. However, and most importantly for continual learning, the performance gains in ES is accompanied by significant forgetting of prior abilities, limiting its applicability for training models online. We also explore the reason behind this behavior and show that the updates made using ES are much less sparse and have orders of magnitude larger $\ell_2$ norm compared to corresponding GRPO updates, explaining the contrasting forgetting curves between the two algorithms. With this study, we aim to highlight the issue of forgetting in gradient-free algorithms like ES and hope to inspire future work to mitigate these issues.

Large language models (LLMs) can be benchmark-contaminated, resulting in inflated scores that mask memorization as generalization, and in multilingual settings, this memorization can even transfer to "uncontaminated" languages. Using the FLORES-200 translation benchmark as a diagnostic, we study two 7-8B instruction-tuned multilingual LLMs: Bloomz, which was trained on FLORES, and Llama as an uncontaminated control. We confirm Bloomz's FLORES contamination and demonstrate that machine translation contamination can be cross-directional, artificially boosting performance in unseen translation directions due to target-side memorization. Further analysis shows that recall of memorized references often persists despite various source-side perturbation efforts like paraphrasing and named entity replacement. However, replacing named entities leads to a consistent decrease in BLEU, suggesting an effective probing method for memorization in contaminated models.

Neural Radiance Fields and 3D Gaussian Splatting have advanced novel view synthesis, yet still rely on dense inputs and often degrade at extrapolated views. Recent approaches leverage generative models, such as diffusion models, to provide additional supervision, but face a trade-off between generalization and fidelity: fine-tuning diffusion models for artifact removal improves fidelity but risks overfitting, while fine-tuning-free methods preserve generalization but often yield lower fidelity. We introduce FreeFix, a fine-tuning-free approach that pushes the boundary of this trade-off by enhancing extrapolated rendering with pretrained image diffusion models. We present an interleaved 2D-3D refinement strategy, showing that image diffusion models can be leveraged for consistent refinement without relying on costly video diffusion models. Furthermore, we take a closer look at the guidance signal for 2D refinement and propose a per-pixel confidence mask to identify uncertain regions for targeted improvement. Experiments across multiple datasets show that FreeFix improves multi-frame consistency and achieves performance comparable to or surpassing fine-tuning-based methods, while retaining strong generalization ability.

Although the capabilities of large language models have been increasingly tested on complex reasoning tasks, their long-horizon planning abilities have not yet been extensively investigated. In this work, we provide a systematic assessment of the planning and long-horizon reasoning capabilities of state-of-the-art Large Reasoning Models (LRMs). We propose a novel benchmark based on Sokoban puzzles, intentionally simplified to isolate long-horizon planning from state persistence. Our findings reveal a consistent degradation in planning performance when more than 25 moves are required to reach the solution, suggesting a fundamental constraint on forward planning capacity. We show that equipping LRMs with Planning Domain Definition Language (PDDL) parsing, validation, and solving tools allows for modest improvements, suggesting inherent architectural limitations which might not be overcome by test-time scaling approaches alone.

Tabular data remains a challenging domain for generative models. In particular, the standard Variational Autoencoder (VAE) architecture, typically composed of multilayer perceptrons, struggles to model relationships between features, especially when handling mixed data types. In contrast, Transformers, through their attention mechanism, are better suited for capturing complex feature interactions. In this paper, we empirically investigate the impact of integrating Transformers into different components of a VAE. We conduct experiments on 57 datasets from the OpenML CC18 suite and draw two main conclusions. First, results indicate that positioning Transformers to leverage latent and decoder representations leads to a trade-off between fidelity and diversity. Second, we observe a high similarity between consecutive blocks of a Transformer in all components. In particular, in the decoder, the relationship between the input and output of a Transformer is approximately linear.

Traditional machine learning systems are typically designed for static data distributions, which suffer from catastrophic forgetting when learning from evolving data streams. Class-Incremental Learning (CIL) addresses this challenge by enabling learning systems to continuously learn new classes while preserving prior knowledge. With the rise of pre-trained models (PTMs) such as CLIP, leveraging their strong generalization and semantic alignment capabilities has become a promising direction in CIL. However, existing CLIP-based CIL methods are often scattered across disparate codebases, rely on inconsistent configurations, hindering fair comparisons, reproducibility, and practical adoption. Therefore, we propose C3Box (CLIP-based Class-inCremental learning toolBOX), a modular and comprehensive Python toolbox. C3Box integrates representative traditional CIL methods, ViT-based CIL methods, and state-of-the-art CLIP-based CIL methods into a unified CLIP-based framework. By inheriting the streamlined design of PyCIL, C3Box provides a JSON-based configuration and standardized execution pipeline. This design enables reproducible experimentation with low engineering overhead and makes C3Box a reliable benchmark platform for continual learning research. Designed to be user-friendly, C3Box relies only on widely used open-source libraries and supports major operating systems. The code is available at https://github.com/LAMDA-CL/C3Box.

Despite growing efforts to mitigate unfairness in recommender systems, existing fairness-aware methods typically fix the fairness requirement at training time and provide limited post-training flexibility. However, in real-world scenarios, diverse stakeholders may demand differing fairness requirements over time, so retraining for different fairness requirements becomes prohibitive. To address this limitation, we propose Cofair, a single-train framework that enables post-training fairness control in recommendation. Specifically, Cofair introduces a shared representation layer with fairness-conditioned adapter modules to produce user embeddings specialized for varied fairness levels, along with a user-level regularization term that guarantees user-wise monotonic fairness improvements across these levels. We theoretically establish that the adversarial objective of Cofair upper bounds demographic parity and the regularization term enforces progressive fairness at user level. Comprehensive experiments on multiple datasets and backbone models demonstrate that our framework provides dynamic fairness at different levels, delivering comparable or better fairness-accuracy curves than state-of-the-art baselines, without the need to retrain for each new fairness requirement. Our code is publicly available at https://github.com/weixinchen98/Cofair.

Whereas reinforcement learning has been applied with success to a range of robotic control problems in complex, uncertain environments, reliance on extensive data - typically sourced from simulation environments - limits real-world deployment due to the domain gap between simulated and physical systems, coupled with limited real-world sample availability. We propose a novel method for sim-to-real transfer of reinforcement learning policies, based on a reinterpretation of neural style transfer from image processing to synthesise novel training data from unpaired unlabelled real world datasets. We employ a variational autoencoder to jointly learn self-supervised feature representations for style transfer and generate weakly paired source-target trajectories to improve physical realism of synthesised trajectories. We demonstrate the application of our approach based on the case study of robot cutting of unknown materials. Compared to baseline methods, including our previous work, CycleGAN, and conditional variational autoencoder-based time series translation, our approach achieves improved task completion time and behavioural stability with minimal real-world data. Our framework demonstrates robustness to geometric and material variation, and highlights the feasibility of policy adaptation in challenging contact-rich tasks where real-world reward information is unavailable.

Time series forecasting is a fundamental problem with applications in climate, energy, healthcare, and finance. Many existing approaches require domain-specific feature engineering and substantial labeled data for each task. We introduce PatchFormer, a patch-based time series foundation model that uses hierarchical masked reconstruction for self-supervised pretraining and lightweight adapters for efficient transfer. PatchFormer segments time series into patches and learns multiscale temporal representations with learnable aggregation across temporal scales. Pretraining uses masked patch reconstruction with dynamic masking and objectives that encourage both local accuracy and global consistency, followed by cross-domain knowledge distillation. Experiments on 24 benchmark datasets spanning weather, energy, traffic, finance, and healthcare demonstrate state-of-the-art zero-shot multi-horizon forecasting, reducing mean squared error by 27.3 percent relative to strong baselines while requiring 94 percent less task-specific training data. The model exhibits near log-linear scaling with more pretraining data up to 100 billion points and processes length-512 sequences 3.8x faster than full-sequence transformers.

This paper introduces a novel Deep Researcher architecture designed to generate detailed research reports on complex PhD level topics by addressing the inherent limitations of the Parallel Scaling paradigm. Our system utilizes two key innovations: Sequential Research Plan Refinement via Reflection and a Candidates Crossover algorithm. The sequential refinement process is demonstrated as an efficient method that allows the agent to maintain a centralized Global Research Context, enabling it to look back at current progress, reason about the research plan, and intelligently make changes at runtime. This dynamic adaptation contrasts with parallel approaches, which often suffer from siloed knowledge. The Candidates Crossover algorithm further enhances search efficiency by deploying multiple LLM candidates with varied parameters to explore a larger search space, with their findings synthesized to curate a comprehensive final research response. The process concludes with One Shot Report Generation, ensuring the final document is informed by a unified narrative and high fact density. Powered by the Gemini 2.5 Pro model, our Deep Researcher was evaluated on the DeepResearch Bench, a globally recognized benchmark of 100 doctoral level research tasks. Our architecture achieved an overall score of 46.21, demonstrating superior performance by surpassing leading deep research agents such as Claude Researcher, Nvidia AIQ Research Assistant, Perplexity Research, Kimi Researcher and Grok Deeper Search present on the DeepResearch Bench actively running leaderboard. This performance marginally exceeds our previous work, Static DRA, and reinforces the finding that sequential scaling consistently outperforms the parallel self consistency paradigm.

Autonomous scientific discovery with large language model (LLM)-based agents has recently made substantial progress, demonstrating the ability to automate end-to-end research workflows. However, existing systems largely rely on runtime-centric execution paradigms, repeatedly reading, summarizing, and reasoning over large volumes of scientific literature online. This on-the-spot computation strategy incurs high computational cost, suffers from context window limitations, and often leads to brittle reasoning and hallucination. We propose Idea2Story, a pre-computation-driven framework for autonomous scientific discovery that shifts literature understanding from online reasoning to offline knowledge construction. Idea2Story continuously collects peer-reviewed papers together with their review feedback, extracts core methodological units, composes reusable research patterns, and organizes them into a structured methodological knowledge graph. At runtime, underspecified user research intents are aligned to established research paradigms, enabling efficient retrieval and reuse of high-quality research patterns instead of open-ended generation and trial-and-error. By grounding research planning and execution in a pre-built knowledge graph, Idea2Story alleviates the context window bottleneck of LLMs and substantially reduces repeated runtime reasoning over literature. We conduct qualitative analyses and preliminary empirical studies demonstrating that Idea2Story can generate coherent, methodologically grounded, and novel research patterns, and can produce several high-quality research demonstrations in an end-to-end setting. These results suggest that offline knowledge construction provides a practical and scalable foundation for reliable autonomous scientific discovery.

Foundation models rely on in-context learning for personalized decision making. The limited size of this context window necessitates memory compression and retrieval systems like RAG. These systems however often treat memory as large offline storage spaces, which is unfavorable for embodied agents that are expected to operate under strict memory and compute constraints, online. In this work, we propose MemCtrl, a novel framework that uses Multimodal Large Language Models (MLLMs) for pruning memory online. MemCtrl augments MLLMs with a trainable memory head μthat acts as a gate to determine which observations or reflections to retain, update, or discard during exploration. We evaluate with training two types of μ, 1) via an offline expert, and 2) via online RL, and observe significant improvement in overall embodied task completion ability on μ-augmented MLLMs. In particular, on augmenting two low performing MLLMs with MemCtrl on multiple subsets of the EmbodiedBench benchmark, we observe that μ-augmented MLLMs show an improvement of around 16% on average, with over 20% on specific instruction subsets. Finally, we present a qualitative analysis on the memory fragments collected by μ, noting the superior performance of μaugmented MLLMs on long and complex instruction types.

Modern industrial and service processes generate high-dimensional, non-Gaussian, and contamination-prone data that challenge the foundational assumptions of classical Statistical Process Control (SPC). Heavy tails, multimodality, nonlinear dependencies, and sparse special-cause observations can distort baseline estimation, mask true anomalies, and prevent reliable identification of an in-control (IC) reference set. To address these challenges, we introduce VSCOUT, a distribution-free framework designed specifically for retrospective (Phase I) monitoring in high-dimensional settings. VSCOUT combines an Automatic Relevance Determination Variational Autoencoder (ARD-VAE) architecture with ensemble-based latent outlier filtering and changepoint detection. The ARD prior isolates the most informative latent dimensions, while the ensemble and changepoint filters identify pointwise and structural contamination within the determined latent space. A second-stage retraining step removes flagged observations and re-estimates the latent structure using only the retained inliers, mitigating masking and stabilizing the IC latent manifold. This two-stage refinement produces a clean and reliable IC baseline suitable for subsequent Phase II deployment. Extensive experiments across benchmark datasets demonstrate that VSCOUT achieves superior sensitivity to special-cause structure while maintaining controlled false alarms, outperforming classical SPC procedures, robust estimators, and modern machine-learning baselines. Its scalability, distributional flexibility, and resilience to complex contamination patterns position VSCOUT as a practical and effective method for retrospective modeling and anomaly detection in AI-enabled environments.

Orthogonal Time Frequency Space (OTFS) modulation offers robust performance in high-mobility scenarios by transforming time-varying channels into the delay-Doppler (DD) domain. However, in high-mobility environment such as emerging 5G Non-Terrestrial Networks (NTN), the extreme orbital velocities of Low Earth Orbit (LEO) satellites frequently cause the physical Doppler shifts to exceed the fundamental grid range. This Doppler ambiguity induces severe model mismatch and renders traditional MLE channel estimators ineffective. To address this challenge, this paper proposes a novel low-complexity pilot-aided Doppler ambiguity detection and compensation framework. We first mathematically derive the OTFS input-output relationship in the presence of aliasing, revealing that Doppler ambiguity manifests itself as a distinct phase rotation along the delay dimension. Leveraging this insight, we developed a two-stage estimator that utilizes pairwise phase differences between pilot symbols to identify the integer ambiguity, followed by a refined Maximum Likelihood Estimation (MLE) for channel recovery. We investigate two pilot arrangements, Embedded Pilot with Guard Zone (EP-GZ) and Data-Surrounded Pilot (DSP), to analyze the trade-off between interference suppression and spectral efficiency. Simulation results demonstrate that the proposed scheme effectively eliminates the error floor caused by ambiguity, achieving Bit Error Rate (BER) and Normalized Mean Square Error (NMSE) performance comparable to the exhaustive search benchmark while maintaining a computational complexity similar to standard MLE.

The construction of Maximum Distance Profile (MDP) convolutional codes in general requires the use of very large finite fields. In contrast convolutional codes with optimal column distances maximize the column distances for a given arbitrary finite field. In this paper, we present a construction of such convolutional codes. In addition, we prove that for the considered parameters the codes that we constructed are the only ones achieving optimal column distances. The structure of the presented convolutional codes with optimal column distances is strongly related to first order Reed-Muller block codes and we leverage this fact to develop a reduced complexity version of the Viterbi algorithm for these codes.

We consider a wireless network comprising multiple singleantenna repeaters that amplify and instantaneously re-transmit received signals in a full-duplex (FD) communication setting. Specifically, we study a massive multiple-input multiple output base station that simultaneously serves multiple uplink (UL) and downlink (DL) user equipment (UE) over the same frequency band. The focus is on the problem of repeater weight optimization at each active repeater to maximize the sum of the weighted minimum spectral efficiencies (SEs) for both UL and DL UEs. The resulting non-convex optimization problem is tackled using a successive convex approximation technique. To demonstrate the effectiveness of the proposed approach, we evaluate its performance against benchmark systems with and without repeater assistance. The optimized FD design achieves SE improvements of up to 4-fold and 2.5-fold compared to its half-duplex counterpart.

Machine learning predictions are increasingly used to supplement incomplete or costly-to-measure outcomes in fields such as biomedical research, environmental science, and social science. However, treating predictions as ground truth introduces bias while ignoring them wastes valuable information. Prediction-Powered Inference (PPI) offers a principled framework that leverages predictions from large unlabeled datasets to improve statistical efficiency while maintaining valid inference through explicit bias correction using a smaller labeled subset. Despite its potential, the growing PPI variants and the subtle distinctions between them have made it challenging for practitioners to determine when and how to apply these methods responsibly. This paper demystifies PPI by synthesizing its theoretical foundations, methodological extensions, connections to existing statistics literature, and diagnostic tools into a unified practical workflow. Using the Mosaiks housing price data, we show that PPI variants produce tighter confidence intervals than complete-case analysis, but that double-dipping, i.e. reusing training data for inference, leads to anti-conservative confidence intervals and coverages. Under missing-not-at-random mechanisms, all methods, including classical inference using only labeled data, yield biased estimates. We provide a decision flowchart linking assumption violations to appropriate PPI variants, a summary table of selective methods, and practical diagnostic strategies for evaluating core assumptions. By framing PPI as a general recipe rather than a single estimator, this work bridges methodological innovation and applied practice, helping researchers responsibly integrate predictions into valid inference.

Large Language Models (LLMs) excel at code generation but struggle with complex problems. Retrieval-Augmented Generation (RAG) mitigates this issue by integrating external knowledge, yet retrieval models often miss relevant context, and generation models hallucinate with irrelevant data. We propose Programming Knowledge Graph (PKG) for semantic representation and fine-grained retrieval of code and text. Our approach enhances retrieval precision through tree pruning and mitigates hallucinations via a re-ranking mechanism that integrates non-RAG solutions. Structuring external data into finer-grained nodes improves retrieval granularity. Evaluations on HumanEval and MBPP show up to 20% pass@1 accuracy gains and a 34% improvement over baselines on MBPP. Our findings demonstrate that our proposed PKG approach along with re-ranker effectively address complex problems while maintaining minimal negative impact on solutions that are already correct without RAG. The replication package is published at https://github.com/iamshahd/ProgrammingKnowledgeGraph

This study provides the first comprehensive empirical mapping of how organizational structures and research portfolios co-occur across U.S. Library and Information Science (LIS) schools. Analyzing 14,705 publications from 1,264 faculty members across 44 institutions (2013--2024), we employ computational methods including word embeddings and topic modeling to identify 16 distinct research themes organized into three foundational dimensions: Library and Knowledge Organization (LKO), Human-Centered Technology (HCT), and Computing Systems (CS). Our mixed-method analysis reveals significant differences in research composition across organizational types: Computer-affiliated schools cluster tightly in computationally-intensive research and differ significantly from all other school types, while independent Information schools demonstrate the greatest research diversity. Temporal analysis of LIS schools reveals complex evolutionary dynamics: 51.4% are moving toward HCT, 37.8% toward CS, and 37.8% toward LKO, with many schools simultaneously shifting along multiple dimensions. Contrary to narratives of computational dominance, HCT emerged as LIS's primary growth vector. These patterns challenge assumptions about field fragmentation, revealing structured diversification shaped by but not determined by organizational positioning. The study provides empirical foundations for institutional strategic planning, accreditation policy, and understanding LIS's evolving disciplinary identity amid computational transformation.

We propose a systematic framework for constructing geometric integrators for Hamiltonian systems on Jacobi manifolds. By combining Poissonization of Jacobi structures with homogeneous symplectic bi-realizations, Jacobi dynamics are lifted to homogeneous Poisson Hamiltonian systems, enabling the construction of structure-preserving Jacobi Hamiltonian integrators. The resulting schemes are constructed explicitly and applied to a range of examples, including contact Hamiltonian systems and classical models. Numerical experiments highlight their qualitative advantages over standard integrators, including better preservation of geometric structure and improved long-time behavior.

This paper presents a comprehensive methodology for implementing knowledge graphs in ROS 2 systems, aiming to enhance the efficiency and intelligence of autonomous robotic missions. The methodology encompasses several key steps: defining initial and target conditions, structuring tasks and subtasks, planning their sequence, representing task-related data in a knowledge graph, and designing the mission using a high-level language. Each step builds on the previous one to ensure a cohesive process from initial setup to final execution. A practical implementation within the Aerostack2 framework is demonstrated through a simulated search and rescue mission in a Gazebo environment, where drones autonomously locate a target. This implementation highlights the effectiveness of the methodology in improving decision-making and mission performance by leveraging knowledge graphs.

Privacy policies are supposed to provide notice. But what if substantive information appears only where users skip it? We identify a structural pattern we call jurisdiction-siloed disclosure: information about data practices appearing in specific, actionable form only within regional compliance sections labeled "California Residents" or "EU/UK Users," while general sections use vague or qualified language for the same practices. Our audit of 123 major companies identifies 282 potential instances across 77 companies (62.6% of this purposive sample). A conservative estimate restricted to practice categories validated against OPP-115 human annotations finds 138 instances across 54 companies (44%); post-2018 categories central to our findings await independent validation. If users skip jurisdiction-labeled sections as information foraging theory predicts, users outside regulated jurisdictions would receive less specific information about practices affecting them--a transparency failure operating through document architecture rather than omission. We propose universal substantive disclosure: practices affecting all users should appear in the main policy body, with regional sections containing only procedural rights information. This standard finds support in analogous disclosure regimes (securities, truth-in-lending, nutritional labeling) where material information must reach all affected parties. Regulators could operationalize this through the FTC's "clear and conspicuous" standard and GDPR transparency principles. This work is hypothesis-generating: we establish that the structural pattern exists and ground the transparency concern in behavioral theory, but direct measurement of jurisdiction-specific section skipping remains the critical validation priority. We release our methodology and annotated dataset to enable replication.

Text-to-video (T2V) diffusion models have achieved rapid progress, yet their demographic biases, particularly gender bias, remain largely unexplored. We present FairT2V, a training-free debiasing framework for text-to-video generation that mitigates encoder-induced bias without finetuning. We first analyze demographic bias in T2V models and show that it primarily originates from pretrained text encoders, which encode implicit gender associations even for neutral prompts. We quantify this effect with a gender-leaning score that correlates with bias in generated videos. Based on this insight, FairT2V mitigates demographic bias by neutralizing prompt embeddings via anchor-based spherical geodesic transformations while preserving semantics. To maintain temporal coherence, we apply debiasing only during early identity-forming steps through a dynamic denoising schedule. We further propose a video-level fairness evaluation protocol combining VideoLLM-based reasoning with human verification. Experiments on the modern T2V model Open-Sora show that FairT2V substantially reduces demographic bias across occupations with minimal impact on video quality.

Federated Learning (FL) enables collaborative training of Large Language Models (LLMs) across distributed data sources while preserving privacy. However, when federated LLMs are deployed in critical applications, it remains unclear which client(s) contributed to specific generated responses, hindering debugging, malicious client identification, fair reward allocation, and trust verification. We present ProToken, a novel Provenance methodology for Token-level attribution in federated LLMs that addresses client attribution during autoregressive text generation while maintaining FL privacy constraints. ProToken leverages two key insights to enable provenance at each token: (1) transformer architectures concentrate task-specific signals in later blocks, enabling strategic layer selection for computational tractability, and (2) gradient-based relevance weighting filters out irrelevant neural activations, focusing attribution on neurons that directly influence token generation. We evaluate ProToken across 16 configurations spanning four LLM architectures (Gemma, Llama, Qwen, SmolLM) and four domains (medical, financial, mathematical, coding). ProToken achieves 98% average attribution accuracy in correctly localizing responsible client(s), and maintains high accuracy when the number of clients are scaled, validating its practical viability for real-world deployment settings.

Forecasting rare events in multivariate time-series data is challenging due to severe class imbalance, long-range dependencies, and distributional uncertainty. We introduce EVEREST, a transformer-based architecture for probabilistic rare-event forecasting that delivers calibrated predictions and tail-aware risk estimation, with auxiliary interpretability via attention-based signal attribution. EVEREST integrates four components: (i) a learnable attention bottleneck for soft aggregation of temporal dynamics; (ii) an evidential head for estimating aleatoric and epistemic uncertainty via a Normal--Inverse--Gamma distribution; (iii) an extreme-value head that models tail risk using a Generalized Pareto Distribution; and (iv) a lightweight precursor head for early-event detection. These modules are jointly optimized with a composite loss (focal loss, evidential NLL, and a tail-sensitive EVT penalty) and act only at training time; deployment uses a single classification head with no inference overhead (approximately 0.81M parameters). On a decade of space-weather data, EVEREST achieves state-of-the-art True Skill Statistic (TSS) of 0.973/0.970/0.966 at 24/48/72-hour horizons for C-class flares. The model is compact, efficient to train on commodity hardware, and applicable to high-stakes domains such as industrial monitoring, weather, and satellite diagnostics. Limitations include reliance on fixed-length inputs and exclusion of image-based modalities, motivating future extensions to streaming and multimodal forecasting.

Theorem proving is fundamental to program verification, where the automated proof of Verification Conditions (VCs) remains a primary bottleneck. Real-world program verification frequently encounters hard VCs that existing Automated Theorem Provers (ATPs) cannot prove, leading to a critical need for extensive manual proofs that burden practical application. While Neural Theorem Proving (NTP) has achieved significant success in mathematical competitions, demonstrating the potential of machine learning approaches to formal reasoning, its application to program verification--particularly VC proving--remains largely unexplored. Despite existing work on annotation synthesis and verification-related theorem proving, no benchmark has specifically targeted this fundamental bottleneck: automated VC proving. This work introduces Neural Theorem Proving for Verification Conditions (NTP4VC), presenting the first real-world multi-language benchmark for this task. From real-world projects such as Linux and Contiki-OS kernel, our benchmark leverages industrial pipelines (Why3 and Frama-C) to generate semantically equivalent test cases across formal languages of Isabelle, Lean, and Rocq. We evaluate large language models (LLMs), both general-purpose and those fine-tuned for theorem proving, on NTP4VC. Results indicate that although LLMs show promise in VC proving, significant challenges remain for program verification, highlighting a large gap and opportunity for future research.

In 2025, coding agents have seen a very rapid adoption. Coding agents leverage Large Language Models (LLMs) in ways that are markedly different from LLM-based code completion, making their study critical. Moreover, unlike LLM-based completion, coding agents leave visible traces in software repositories, enabling the use of MSR techniques to study their impact on SE practices. This paper documents the promises, perils, and heuristics that we have gathered from studying coding agent activity on GitHub.

Foundation models like the Segment Anything Model (SAM) show strong generalization, yet adapting them to medical images remains difficult due to domain shift, scarce labels, and the inability of Parameter-Efficient Fine-Tuning (PEFT) to exploit unlabeled data. While conventional models like U-Net excel in semi-supervised medical learning, their potential to assist a PEFT SAM has been largely overlooked. We introduce SC-SAM, a specialist-generalist framework where U-Net provides point-based prompts and pseudo-labels to guide SAM's adaptation, while SAM serves as a powerful generalist supervisor to regularize U-Net. This reciprocal guidance forms a bidirectional co-training loop that allows both models to effectively exploit the unlabeled data. Across prostate MRI and polyp segmentation benchmarks, our method achieves state-of-the-art results, outperforming other existing semi-supervised SAM variants and even medical foundation models like MedSAM, highlighting the value of specialist-generalist cooperation for label-efficient medical image segmentation. Our code is available at https://github.com/vnlvi2k3/SC-SAM.

Diffusion Large Language Models (dLLMs) offer fast, parallel token generation, but their standalone use is plagued by an inherent efficiency-quality tradeoff. We show that, if carefully applied, the attributes of dLLMs can actually be a strength for drafters in speculative decoding with autoregressive (AR) verifiers. Our core insight is that dLLM's speed from parallel decoding drastically lowers the risk of costly rejections, providing a practical mechanism to effectively realize the (elusive) lengthy drafts that lead to large speedups with speculative decoding. We present FailFast, a dLLM-based speculative decoding framework that realizes this approach by dynamically adapting its speculation length. It "fails fast" by spending minimal compute in hard-to-speculate regions to shrink speculation latency and "wins big" by aggressively extending draft lengths in easier regions to reduce verification latency (in many cases, speculating and accepting 70 tokens at a time!). Without any fine-tuning, FailFast delivers lossless acceleration of AR LLMs and achieves up to 4.9$\times$ speedup over vanilla decoding, 1.7$\times$ over the best naive dLLM drafter, and 1.7$\times$ over EAGLE-3 across diverse models and workloads. We open-source FailFast at https://github.com/ruipeterpan/failfast.

We present ReactionMamba, a novel framework for generating long 3D human reaction motions. Reaction-Mamba integrates a motion VAE for efficient motion encoding with Mamba-based state-space models to decode temporally consistent reactions. This design enables ReactionMamba to generate both short sequences of simple motions and long sequences of complex motions, such as dance and martial arts. We evaluate ReactionMamba on three datasets--NTU120-AS, Lindy Hop, and InterX--and demonstrate competitive performance in terms of realism, diversity, and long-sequence generation compared to previous methods, including InterFormer, ReMoS, and Ready-to-React, while achieving substantial improvements in inference speed.