We present a novel theoretical framework, Q-MMR, for off-policy evaluation in finite-horizon MDPs. Q-MMR learns a set of scalar weights, one for each data point, such that the reweighted rewards approximate the expected return under the target policy. The weights are learned inductively in a top-down manner via a moment matching objective against a value-function discriminator class. Notably, and perhaps surprisingly, a data-dependent finite-sample guarantee for general function approximation can be established under only the realizability of $Q^π$, with a dimension-free bound -- that is, the error does not depend on the statistical complexity of the function class. We also establish connections to several existing methods, such as importance sampling and linear FQE. Further theoretical analyses shed new light on the nature of coverage, a concept of fundamental importance to offline RL.

In a conversation, a helpful assistant must reliably follow user directives, even as they refine, modify, or contradict earlier requests. Yet most instruction-following benchmarks focus on single-turn or short multi-turn scenarios, leaving open how well models handle long-horizon instruction-following tasks. To bridge this gap, we present SEQUOR, an automatic benchmark for evaluating constraint adherence in long multi-turn conversations. SEQUOR consists of simulated persona-driven interactions built with constraints extracted from real-world conversations. Our results show that even when following a single constraint, instruction-following accuracy consistently decreases as the conversation grows longer, with drops exceeding 11%. This decline becomes larger when models have to follow multiple constraints simultaneously, reducing their accuracy by over 40%. In scenarios where constraints are added or replaced at arbitrary points of the conversation, model accuracy decreases by more than 9%. Taken together, our results reveal that current models still struggle to follow user instructions in multi-turn conversations, and provide a way for better measuring instruction-following capabilities in assistants.

Diabetic Retinopathy (DR) is a leading cause of preventable blindness among working-age adults worldwide, yet most automated screening systems are limited to image-level classification and lack clinically structured reporting. We propose Retina-RAG, a low-cost modular framework that jointly performs DR severity grading, macular edema (ME) detection, and report generation. The architecture decouples a high-performance retinal classifier and a parameter-efficient vision-language model (Qwen2.5-VL-7B-Instruct) adapted via Low-Rank Adaptation (LoRA), enabling flexible component integration. A retrieval-augmented generation (RAG) module injects curated ophthalmic knowledge together with structured classifier outputs at inference time to improve diagnostic consistency and reduce hallucinations. Retina-RAG achieves an F1-score of 0.731 for DR grading and 0.948 for ME detection, substantially outperforming zero-shot Qwen (0.096, 0.732) and MMed-RAG (0.541, 0.641) on a retinal disease detection dataset with captions. For report generation, Retina-RAG attains ROUGE-L 0.438 and SBERT similarity 0.884, exceeding all baselines. The full framework operates on a single consumer-grade GPU, demonstrating that clinically structured retinal AI can be achieved with modest computational resources.

Agentic systems increasingly solve complex user requests by executing orchestrated workflows, where subtasks are assigned to specialized models or tools and coordinated according to their dependencies. While recent work improves agent efficiency by optimizing the performance--cost--latency frontier, real deployments often impose concrete requirements: a workflow must be completed within a specified budget and before a specified deadline. This shifts the goal from average efficiency optimization to maximizing the probability that the entire workflow completes successfully under explicit budget and deadline constraints. We study \emph{constraint-driven online resource allocation for agentic workflows}. Given a dependency-structured workflow and estimates of success rates and generation lengths for each subtask--model pair, the executor dynamically allocates models and parallel samples across simultaneously executable subtasks while managing the remaining budget and time. We formulate this setting as a finite-horizon stochastic online allocation problem and propose \emph{Monte Carlo Portfolio Planning} (MCPP), a lightweight closed-loop planner that directly estimates constrained completion probability through simulated workflow executions and replans after observed outcomes. Experiments on CodeFlow and ProofFlow demonstrate that MCPP consistently improves constrained completion probability over strong baselines across a wide range of budget--deadline constraints.

LLMs reliably correct false claims when presented in isolation, yet when the same claims are embedded in task-oriented requests, they often comply rather than correct. We term this failure mode \emph{correction suppression} and construct a benchmark of 300 false premises to systematically evaluate it across eight models. Suppression rates range from 19\% to 90\%, with four models exceeding 80\%, establishing correction suppression as a prevalent and severe phenomenon. Mechanistic analysis reveals that suppression is not a knowledge failure: the model registers the error internally but task context diverts early-layer attention from the false claim as output intent crystallizes toward compliance at middle layers. We characterize this as \emph{knowing but not correcting} -- suppression occurs at response selection rather than knowledge encoding. Guided by this mechanism, we propose two training-free interventions. Correction Direction Steering (CDS) estimates a correction-compliance direction from matched pairs and injects it at middle layers before output intent crystallizes. Dynamic Payload Amplification (DPA) localizes payload tokens via attention divergence between early and late layers and amplifies their representation at the final layer, requiring no calibration data. Experiments on Qwen3.5-9B and LLaMA3.1-8B show both methods substantially improve factual strictness. CDS achieves the highest correction rate on Qwen3.5-9B (0\%$\to$58.2\%). DPA is the only method that preserves or improves reasoning capability on both models. These findings introduce \emph{factual strictness} -- the willingness to uphold accuracy against contextual pressures -- as a new dimension of model reliability.

Video large multimodal models increasingly face a scalability bottleneck: long videos produce excessively long visual-token sequences, which sharply increase memory and latency during inference. While existing compression methods are effective in specific settings, most are either weakly query-aware or apply a fixed compression policy across frames, proving suboptimal when visual evidence is unevenly distributed over time. To address this, we present VideoRouter, a query-adaptive dual-router framework built on InternVL for budgeted evidence allocation. The Semantic Router predicts the dominant allocation policy, choosing between broad temporal coverage and adaptive high-resolution preservation, while the Image Router uses early LLM layers to score frame relevance. This enables aggressive compression on less relevant frames while preserving detail on critical evidence frames. To train both routers, we build Video-QTR-10K for allocation-policy supervision and Video-FLR-200K for frame-relevance supervision. Experiments on VideoMME, MLVU, and LongVideoBench show that VideoRouter consistently improves over the InternVL baseline under comparable or lower budgets, achieving up to a 67.9% token reduction.

Conformal prediction is a framework that provides valid uncertainty quantification for general models with exchangeable data. However, in the online learning and time-series settings, exchangeability is not satisfied. Existing online conformal methods, such as adaptive conformal inference (ACI), can achieve long-run validity, yet they remain inefficient under covariate heterogeneity because they rely on global calibration. We propose \emph{Online Localized Conformal Prediction (OLCP)}, which combines online adaptation with covariate-dependent localization to better reflect heterogeneity. To reduce sensitivity to the localization bandwidth, we further develop \emph{OLCP-Hedge}, which performs bandwidth selection as an online expert aggregation problem using a constrained online convex optimization framework. Importantly, we provide coverage guarantees for both algorithms and demonstrate through simulations and real-data experiments that the proposed methods attain valid long-run coverage with narrower prediction sets than existing baselines.

Designing reward functions for agile robotic maneuvers in reinforcement learning remains difficult, and demonstration-based approaches often require reference motions that are unavailable for novel platforms or extreme stunts. We present LineRides, a line-guided learning framework that enables a custom bicycle robot to acquire diverse, commandable stunt behaviors from a user-provided spatial guideline and sparse key-orientations, without demonstrations or explicit timing. LineRides handles physically infeasible guidelines using a tracking margin that permits controlled deviation, resolves temporal ambiguity by measuring progress via traveled distance along the guideline, and disambiguates motion details through position- and sequence-based key-orientations. We evaluate LineRides on the Ultra Mobility Vehicle (UMV) and show that the policy trained with our methods supports seamless transitions between normal driving and stunt execution, enabling five distinct stunts on command: MiniHop, LargeHop, ThreePointTurn, Backflip, and DriftTurn.

In deep reinforcement learning (DRL), an agent is trained from a stream of experience. In a continual learning setting, such agents can suffer from plasticity loss: their ability to learn new skills from new experiences diminishes over training. Recently, Mixture-of-Experts (MoE) networks have been reported to enable scaling laws and facilitate the learning of diverse skills. However, in continual reinforcement learning settings, their performance can degenerate as learning proceeds, indicating a loss of plasticity. To address this, building on Neural Tangent Kernel (NTK) theory, we formalize the plasticity loss in MoE policies as a loss of spectral plasticity. We then derive a tractable proxy for spectral plasticity, one expressible in terms of individual expert feature matrices. Leveraging this proxy, we introduce SPHERE, a practical Parseval penalty tailored for MoE-based policies that alleviates the loss of spectral plasticity. On MetaWorld and HumanoidBench, SPHERE improves average success under continual RL by 133% and 50% over an unregularized MoE baseline, while maintaining higher spectral plasticity throughout training.

Adapting pretrained models typically involves a trade-off between the high training costs of backpropagation and the heavy inference overhead of memory-based or in-context learning. We propose FAAST, a forward-only associative adaptation method that analytically compiles labeled examples into fast weights in a single pass. By eliminating memory or context dependence, FAAST achieves constant-time inference and decouples task adaptation from pretrained representation. Across image classification and language modeling benchmarks, FAAST matches or exceeds backprop-based adaptation while reducing adaptation time by over 90% and is competitive to memory/context-based adaptation while saving memory usage by up to 95%. These results demonstrate FAAST as a highly efficient, scalable solution for supervised task adaptation, particularly for resource-constrained models. We release the code and models at https://github.com/baoguangsheng/faast.

Transportation surveys are widely used to understand travel preferences and adoption barriers, yet most survey-based analyses remain descriptive or predictive and rarely provide sparse, policy-feasible intervention strategies. We study sparse counterfactual community intervention from survey responses, where the goal is to shift a target respondent group toward a desired reference group through controllable survey-variable adjustments. We formulate this task as a policy-feasible distributional alignment problem using a fixed-basis nonnegative latent representation that preserves pre/post comparability and provides a stable map from latent factors to original variables. To make latent movement actionable, target-relevant latent factors are identified through Shapley-guided attribution and transferred to controllable variables as intervention priorities. Feasible group-level adjustments are then learned by minimizing an entropy-regularized optimal-transport discrepancy between the post-intervention target distribution and the reference distribution, together with a weighted $\ell_{2,1}$ penalty that promotes shared policy-lever sparsity. Experiments on real-world transportation survey datasets show that the proposed framework produces compact and interpretable policy-feasible interventions with explicit adjustment magnitudes, improves population-level conversion, and preserves intervention sparsity. Code and datasets are publicly available at: https://github.com/pangjunbiao/latent-group-alignment.git

Understanding soil is fundamental to agriculture, carbon cycling, and environmental sustainability, yet progress is limited by fragmented and heterogeneous datasets that constrain modeling to small-scale predictive settings rather than high-dimensional representation learning. We introduce LUCAS-MEGA, a large-scale multimodal dataset constructed through systematic data fusion of European soil-environment observations, with the LUCAS survey as its backbone. The fused dataset comprises over 70,000 samples and more than 1,000 features spanning physical, chemical, environmental, biological, and visual attributes, aggregated from 68 source datasets. To enable integration at scale, we develop SoilFuser, a multi-agent, human-in-the-loop data fusion pipeline that standardizes heterogeneous data formats and measurement protocols, resolves inconsistencies and invalid entries (e.g., unit inconsistencies, codebook mismatches, and erroneous values), incorporates natural language annotations, and harmonizes multimodal attributes and metadata into a unified, machine learning-ready feature space. The resulting dataset captures key characteristics of real-world soil observations, including multimodality, uneven feature coverage, and heterogeneous uncertainty. To demonstrate the usability of LUCAS-MEGA for data-driven modeling, we pretrain a multimodal tabular transformer (SoilFormer) using a self-supervised objective based on feature masking, achieving stable training, strong predictive performance, and representations that support uncertainty-aware prediction. We further show that the learned representations recover relationships consistent with established soil processes. LUCAS-MEGA is released with open access and is accompanied by composable, agent-friendly APIs that support structured querying and data-driven workflows.

Causal Bayesian Networks (CBNs) are a powerful tool for reasoning under uncertainty about complex real-world problems. Such problems evolve over time, responding to external shocks as they occur. To support decision-making, CBNs require a cause-and-effect map of the variables under consideration, known as the network's structure. Learning the graphical structure of a causal model from data remains challenging; learning it from time-series data is even harder because dependencies may arise at different time lags. Existing time-series causal discovery methods often assume a fixed lag window and do not explicitly optimise edge-specific lags. We propose a Tabu-based structure learning algorithm that searches for a time-ordered directed structure (i.e., where every edge respects time) while allowing edge-specific lags up to a specified maximum lag. The approach uses a decomposable BIC-based score with node-specific effective sample sizes and an explicit lag-length penalty encouraging parsimonious delay assignments while preserving efficient local score updates. We provide theoretical guarantees of validity and local optimality, and we also describe a parallel implementation for improved scalability. In simulations, the method recovered graph structure competitively and estimated lags accurately when true adjacencies were recovered. On a real-world UK COVID-19 policy dataset, the learnt structure was dominated by short delays while retaining a substantial minority of longer-lag dependencies, consistent with delayed behavioural and epidemiological effects.

We propose HeadsUp, a scalable feed-forward method for reconstructing high-quality 3D Gaussian heads from large-scale multi-camera setups. Our method employs an efficient encoder-decoder architecture that compresses input views into a compact latent representation. This latent representation is then decoded into a set of UV-parameterized 3D Gaussians anchored to a neutral head template. This UV representation decouples the number of 3D Gaussians from the number and resolution of input images, enabling training with many high-resolution input views. We train and evaluate our model on an internal dataset with more than 10,000 subjects, which is an order of magnitude larger than existing multi-view human head datasets. HeadsUp achieves state-of-the-art reconstruction quality and generalizes to novel identities without test-time optimization. We extensively analyze the scaling behavior of our model across identities, views, and model capacity, revealing practical insights for quality-compute trade-offs. Finally, we highlight the strength of our latent space by showcasing two downstream applications: generating novel 3D identities and animating the 3D heads with expression blendshapes.

Reinforcement learning is crucial for aligning large language models to perform complex reasoning tasks. However, current algorithms such as Group Relative Policy Optimization suffer from coarse grained, sequence level credit assignment, which severely struggles to isolate pivotal reasoning steps within long Chain of Thought generations. Furthermore, the standard unbounded Kullback Leibler divergence penalty induces severe gradient instability and mode seeking conservatism, ultimately stifling the discovery of novel reasoning trajectories. To overcome these limitations, we introduce Distribution Guided Policy Optimization, a novel critic free reinforcement learning framework that reinterprets distribution deviation as a guiding signal rather than a rigid penalty. DGPO replaces the volatile KL divergence with the bounded Hellinger distance to safely quantify token level exploration without the risk of gradient explosion. To effectively distinguish genuine reasoning breakthroughs from hallucinatory noise, we propose an entropy gating mechanism that scales this deviation by the policy`s epistemic uncertainty. By dynamically redistributing the coarse sequence-level advantage to individual tokens based on these gated scores, DGPO heavily incentivizes critical exploratory steps while suppressing unwarranted, low-entropy deviations. Consequently, DGPO completely eliminates the traditional token-level KL penalty and achieves fine-grained credit reallocation without the computational overhead of an additional value network. Extensive empirical evaluations demonstrate that DGPO sets a new state-of-the-art for critic free alignment. Notably, on the Qwen2.5-32B architecture, DGPO achieves 60.0% Avg@32 accuracy and 46.0% Avg@32 accuracy on the challenging AIME2024 and AIME2025 benchmarks respectively, substantially outperforming competitive baselines like DAPO.

In 3D point cloud understanding, the core challenge lies in accurately capturing discriminative features within complex neighborhoods, which directly affects the execution precision of downstream tasks such as embodied AI and autonomous driving. Existing methods explore feature correlation discrimination but are limited to point-level spatial distribution or channel responses, enabling only coarse-grained level evaluation. For modern multi-scale point cloud networks, such coarse-grained metrics inevitably incur significant information loss in deeper layers. To address this, we propose PointCRA, a novel network with a channel-level metric-based enhancement mechanism. Our core idea is to introduce temporal trend variation as a new evaluation dimension to avoid the information loss caused by weight dimension collapse in existing spatial and channel attention mechanisms. On this basis, we construct a multi-level calibration framework guided by neighborhood homogeneity for weight calibration, and design a dedicated loss function to enhance channel discriminability.PointCRA leverages intrinsic feature priors to adaptively correct feature aggregation, offering interpretability with low parameter overhead. Our method is transferable, interpretable, and efficient. We validate the proposed method on diverse datasets and benchmark models, and further demonstrate its rationality through extensive analytical experiments. Our PointCRA achieves 77.5\% mIoU on the S3DIS dataset, 90.4\% OA on the ScanObjectNN dataset, and 87.4\% instance mIoU on the ShapeNetPart dataset. The code and pretrained weights are publicly available on GitHub: https://github.com/AGENT9717/PointCRA

Machine unlearning aims to remove specified training data to satisfy privacy regulations such as GDPR. However, existing evaluations assume identical precision at unlearning and deployment, overlooking that production LLMs are deployed at low-bit precision. We show that INT4 quantization systematically restores forgotten content even when models pass compliance audits at bfloat16 (BF16), we term this the quantization recovery attack (QRA). We conduct the first systematic study of unlearning robustness under adapter-space INT4 quantization in the NF4+LoRA regime, evaluating seven methods on LLaMA-3-8B-Instruct across TOFU, MUSE-News, and WikiBio-WPU. INT8 is benign; INT4 induces recovery of up to 22x, worsening with dataset difficulty. We identify the FA-RA-Q-INT4 trilemma: no method simultaneously achieves strong forgetting, high utility, and quantization robustness. A dense Pareto sweep reveals a sharp phase transition once robustness is achieved, retaining accuracy collapses regardless of further tuning. To address this, we propose DURABLEUN-SAF (Sharpness-Aware Forgetting), a quantization-aware objective using Straight-Through Estimator gradients through INT4 rounding. DURABLEUN-SAF is the only method to achieve a stable empirical (0.047, {BF16, INT8, INT4})- durability certificate: Q-INT4= 0.043 +- 0.002, cert rate= 3/3, versus SalUn's cert rate= 1/3 at its own published hyperparameters. We call for Q-INT4 to be adopted as a standard evaluation metric alongside FA and RA.

Mathematical reasoning is a key benchmark for large language models. Reinforcement learning is a standard post-training mechanism for improving the reasoning capabilities of large language models, yet performance remains sensitive to the design of the reward function that drives policy optimization. This paper introduces a search-driven framework that treats the reward specification itself as an object of optimization. The setting of interest is one in which the base model is held fixed and the reward specification is the primary remaining design lever. Candidate reward functions are generated by a frontier language model, validated automatically, screened through 500-step Group Relative Policy Optimization (GRPO) training runs on a Llama-3.2-3B-Instruct base model with Low-Rank Adaptation (LoRA), and ranked by F1 on the GSM8K test set. Ranked summaries from prior rounds are then fed back into the next round of generation. Over five rounds, the search produces 50 candidate rewards. The mean F1 rises from 0.596 in Round 1 to 0.632 in Round 5, and the top individual reward reaches F1 = 0.787. Seven ensemble configurations of top-ranked rewards are evaluated. The best ensemble achieves F1 = 0.795 (95% bootstrap CI [0.756, 0.832]) and accuracy 0.660 [0.635, 0.686], a 0.19 absolute F1 gain over a base-rewards-only GRPO baseline (F1 = 0.609). Pairwise McNemar tests with Bonferroni correction show all five-or-more-reward configurations are statistically indistinguishable at α = 0.05/21. A three-seed re-training of the best ensemble yields F1 of 0.785. A randomly drawn 5-reward control collapses to F1 = 0.047, which shows that the ranked-feedback loop, not the additive signal of having more rewards, drives the gain.

Multi-Hop Fact Verification (MHFV) necessitates complex reasoning across disparate evidence, posing significant challenges for Large Language Models (LLMs) which often suffer from hallucinations and fractured logical chains. Existing methods, while improving transparency via Chain-of-Thought (CoT), lack explicit modeling of the causal dependencies between evidence and claims. In this work, we introduce a novel framework that grounds reasoning in a Structural Causal Model (SCM), treating verification as a constructive causal inference process. We empirically identify an "inverted U-shaped" correlation between reasoning chain length and accuracy, revealing that excessive structural complexity degrades performance. To address this, we propose a Rule-based Reinforcement Learning strategy using Group Relative Policy Optimization (GRPO). This approach dynamically optimizes the trade-off between structural depth and conciseness. Extensive experiments on HoVer and EX-FEVER demonstrate that our SCM-GRPO framework significantly outperforms state-of-the-art baselines, offering a reliable and interpretable solution for complex fact verification.

In deep networks with small initialization, training exhibits long plateaus separated by sharp feature-acquisition transitions. Whereas shallow nonlinear networks and deep linear networks are well studied, extending these analyses to deep nonlinear networks remains challenging. We derive an exact identity for the imbalance of Frobenius norms of layer weight matrices that holds for any smooth activation and any differentiable loss and use this to classify activation functions into four universality classes. On the permutation-symmetric submanifold, the identity combines with an approximate balance law to reduce the full matrix flow to a scalar ODE, giving a critical-depth escape time law $τ_\star = Θ(\varepsilon^{-(r-2)})$ governed by the number $r$ of layers at the bottleneck scale rather than the total depth $L$. We find that this same $r-2$ exponent is recovered under He-normal initialization with $r$ bottleneck layers rescaled by $\varepsilon$, where the symmetry manifold is preserved by the flow but not attracting. We find close agreement between our theory and numerical simulations.

While autoregressive Large Vision-Language Models (LVLMs) demonstrate remarkable proficiency in multimodal tasks, they face a "Visual Signal Dilution" phenomenon, where the accumulation of textual history expands the attention partition function, causing visual attention to decay inversely with generated sequence length. To counteract this, we propose Persistent Visual Memory (PVM), a lightweight learnable module designed to strengthen sustained, on-demand access to visual evidence. Integrated as a parallel branch alongside the Feed-Forward Network (FFN) in LVLMs, PVM establishes a distance-agnostic retrieval pathway that directly provides visual embeddings for enhanced visual perception, thereby structurally mitigating the signal suppression inherent to deep generation. Extensive experiments on Qwen3-VL models demonstrate that PVM brings notable improvements with negligible parameter overhead, delivering consistent average accuracy gains across both 4B and 8B scales, particularly in complex reasoning tasks that demand persistent visual perception. Furthermore, in-depth analysis reveals that PVM shows improved robustness in longer generations and accelerates internal prediction convergence.

Reinforcement Learning with Verifiable Rewards (RLVR) enhances reasoning of Large Language Models (LLMs) but usually exhibits limited generation diversity due to the over-incentivization of positive rewards. Although methods like Negative Sample Reinforcement (NSR) mitigate this issue by upweighting penalty from negative samples, they may suppress the semantic distributions shared between positive and negative responses. To boost reasoning ability without losing diversity, this paper proposes negative sample projection Residual Reinforcement Learning (ResRL) that decouples similar semantic distributions among positive and negative responses. We theoretically link Lazy Likelihood Displacement (LLD) to negative-positive head-gradient interference and derive a single-forward proxy that upper-bounds representation alignment to guide conservative advantage reweighting. ResRL then projects negative-token hidden representations onto an SVD-based low-rank positive subspace and uses projection residuals to modulate negative gradients, improving reasoning while preserving diversity and outperforming strong baselines on average across twelve benchmarks spanning Mathematics, Code, Agent Tasks, and Function Calling. Notably, ResRL surpasses NSR on mathematical reasoning by 9.4\% in Avg@16 and 7.0\% in Pass@128. Code is available at https://github.com/1229095296/ResRL.git.

Embodied AI and robotic systems increasingly depend on scalable, diverse, and physically grounded 3D content for simulation-based training and real-world deployment. While 3D generative modeling has advanced rapidly, embodied applications impose requirements far beyond visual realism: generated objects must carry kinematic structure and material properties, scenes must support interaction and task execution, and the resulting content must bridge the gap between simulation and reality. This survey reviews 3D generation for embodied AI and organizes the literature around three roles that 3D generation plays in embodied systems. In Data Generator, 3D generation produces simulation-ready objects and assets, including articulated, physically grounded, and deformable content for downstream interaction; in Simulation Environments, it constructs interactive and task-oriented worlds, spanning structure-aware, controllable, and agentic scene generation; and in Sim2Real Bridge, it supports digital twin reconstruction, data augmentation, and synthetic demonstrations for downstream robot learning and real-world transfer. We also show that the field is shifting from visual realism toward interaction readiness, and we identify the main bottlenecks, including limited physical annotations, the gap between geometric quality and physical validity, fragmented evaluation, and the persistent sim-to-real divide, that must be addressed for 3D generation to become a dependable foundation for embodied intelligence. Our project page is at https://3dgen4robot.github.io.

The rapid growth in the size of large language models has necessitated the partitioning of computational workloads across accelerators such as GPUs, TPUs, and NPUs. However, these parallelization strategies incur substantial data communication overhead significantly hindering computational efficiency. While communication-computation overlap presents a promising direction, existing data slicing based solutions suffer from tail latency. To overcome this limitation, this research introduces a novel communication-computation overlap technique to eliminate this tail latency in state of the art overlap methods for distributed LLM training. The aim of this technique is to effectively mitigate communication bottleneck of tensor parallelism and data parallelism for distributed training and inference. In particular, we propose a novel method termed CommFuse that replaces conventional collective operations of reduce-scatter and all-gather with decomposed peer-to-peer (P2P) communication and schedules partitioned computations to enable fine-grained overlap. Our method provides an exact algorithm for reducing communication overhead that eliminates tail latency. Moreover, it presents a versatile solution compatible with data-parallel training and various tensor-level parallelism strategies, including TPSP and UP. Experimental evaluations demonstrate that our technique consistently achieves lower latency, superior Model FLOPS Utilization (MFU), and high throughput.

Target Safety Assessment (TSA) requires systematic integration of heterogeneous evidence, including genetic, transcriptomic, target homology, pharmacological, and clinical data, to evaluate potential safety liabilities of therapeutic targets. This process is inherently iterative and expert-driven, posing challenges in scalability and reproducibility. We present TSAssistant, a multi-agent framework designed to support TSA report drafting through a modular, section-based, and human-in-the-loop paradigm. The framework decomposes report generation into a coordinated pipeline of specialised subagents, each targeting a single TSA section. Specialised subagents retrieve structured and unstructured data as well as literature evidence from curated biomedical sources through standardised tool interfaces, producing individually citable, evidence-grounded sections. Agent behaviour is governed by a hierarchical instruction architecture comprising system prompts, domain-specific skill modules, and runtime user instructions. A key feature is an interactive refinement loop in which users may manually edit sections, append new information, upload additional sources, or re-invoke agents to revise specific sections, with the system maintaining conversational memory across iterations. TSAssistant is designed to reduce the mechanical burden of evidence synthesis and report drafting, supporting a hybrid model in which agentic AI augments evidence synthesis while toxicologists retain final decision authority.

Fault location in distribution grids is critical for reliability and minimizing outage durations. Yet, it remains challenging due to partial observability, given sparse measurement infrastructure. Recent works show promising results by combining Recurrent Neural Networks (RNNs) and Graph Neural Networks (GNNs) for spatio-temporal learning. Still, many modern GNN architectures remain untested for this grid application, while existing GNN solutions have not explored GNN topology definitions beyond simply adopting the full grid topology to construct the GNN graph. We address these gaps by (i) systematically comparing a newly proposed graph-forming strategy (measured-only) to the traditional full-topology approach, and (ii) introducing STGNN (Spatio-temporal GNN) models based on GraphSAGE and an improved Graph Attention (GATv2), for distribution grid fault location; (iii) benchmarking them against state-of-the-art STGNN and RNN baselines on the IEEE 123-bus feeder. In our experiments, all evaluated STGNN variants achieve high performance and consistently outperform a pure RNN baseline, with improvements up to 11 percentage points F1. Among STGNN models, the newly explored RGATv2 and RGSAGE achieve only marginally higher F1 scores. Still, STGNNs demonstrate superior stability, with tight confidence intervals (within +/- 1.4%) compared to the RNN baseline (up to +/- 7.5%) across different experiment runs. Finally, our proposed reduced GNN topology (measured-only) shows clear benefits in both (i) model training time (6-fold reduction) and (ii) model performance (up to 11 points F1). This suggests that measured-only graphs offer a more practical, efficient, and robust framework for partially observable distribution grids.

Multimodal large language models (MLLMs) have shown promising reasoning abilities, yet evaluating their performance in specialized domains remains challenging. STEM reasoning is a particularly valuable testbed because it provides highly verifiable feedback, but existing benchmarks often permit unimodal shortcuts due to modality redundancy and focus mainly on final-answer accuracy, overlooking the reasoning process itself. To address this challenge, we introduce StepSTEM: a graduate-level benchmark of 283 problems across mathematics, physics, chemistry, biology, and engineering for fine-grained evaluation of cross-modal reasoning in MLLMs. StepSTEM is constructed through a rigorous curation pipeline that enforces strict complementarity between textual and visual inputs. We further propose a general step-level evaluation framework for both text-only chain-of-thought and interleaved image-text reasoning, using dynamic programming to align predicted reasoning steps with multiple reference solutions. Experiments across a wide range of models show that current MLLMs still rely heavily on textual reasoning, with even Gemini 3.1 Pro and Claude Opus 4.6 achieving only 38.29% accuracy. These results highlight substantial headroom for genuine cross-modal STEM reasoning and position StepSTEM as a benchmark for fine-grained evaluation of multimodal reasoning. Source code is available at https://github.com/lll-hhh/STEPSTEM.

Many high-stakes decisions depend on forecasts made before outcomes are known. In this future prediction setting, the central challenge is that public evidence evolves over time, while the main supervision signal arrives only after resolution: the realized outcome mainly assesses final correctness, offering only coarse guidance on what to track, what to verify, and which judgments to leave uncertain along the way. Our key observation is that revisiting the same unresolved question over time creates informative temporal contrasts across evolving evidence and repeated forecasts, exposing what earlier attempts missed before resolution and yielding a diagnostic signal we call the pre-resolution signal. We instantiate this idea in Milkyway, a future prediction agent with a persistent future prediction harness, an editable external state that stores reusable procedural guidance across revisits to the same unresolved question. As the same unresolved question is revisited, Milkyway extracts pre-resolution signals from evolving evidence and repeated forecasts, uses them to update the harness, and improves later forecasts on that question before resolution. After resolution, the realized outcome serves as a post-resolution check of provisional updates. On the FutureX and FutureWorld benchmarks, Milkyway achieves strong performance against competitive baselines, and a mechanism study suggests that the gains stem from harness evolution driven by pre-resolution signals rather than repeated prediction alone.

Drifting models have recently gained attention for generating high-quality samples in a single forward pass. During training, they learn a push-forward map by following a vector-valued field, the drift field. We ask whether this procedure is equivalent to optimizing a scalar loss and find that, in general, it is not: drift fields are not conservative and cannot be written as the gradient of any scalar potential. We identify the position-dependent normalization as the source of non-conservatism, with the Gaussian kernel as the unique radial exception. Guided by this, we introduce the sharp kernel $k^\#$ and a sharp-normalized drift field that is conservative for general radial kernels. The resulting vector field is the gradient of a scalar potential that can be optimized directly using stochastic gradient descent. Moreover, the field has the form of a score difference of kernel density estimates, and gives exact equilibrium identifiability. Thus, sharp normalization closes the gap to related literature, such as Wasserstein gradient-flows and denoising score matching, also for non-Gaussian kernels. Empirically, sharp normalization preserves the performance of the original drifting objective, suggesting that the non-conservative flexibility is not required for high-quality generation.

How do people acquire rich, flexible knowledge about their environment from others despite limited cognitive capacity? Humans are often thought to rely on computationally costly mentalizing, such as inferring others' beliefs. In contrast, cultural evolution emphasizes that behavioral transmission can be supported by simple social cues. Using reinforcement learning simulations, we show how minimal social learning can indirectly transmit higher-level representations. We simulate a naïve agent searching for rewards in a reconfigurable environment, learning either alone or by observing an expert - crucially, without inferring mental states. Instead, the learner heuristically selects actions or boosts value representations based on observed actions. Our results demonstrate that these cues bias the learner's experience, causing its representation to converge toward the expert's. Model-based learners benefit most from social exposure, showing faster learning and more expert-like representations. These findings show how cultural transmission can arise from simple, non-mentalizing processes exploiting asocial learning mechanisms.