YOUR RESEARCH, IN SIGHT
覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。
时间范围跟随排序：版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常，查看状态 →
盛碧云, 徐志颖, 蔡惠, 周剑, 肖甫
现有的基于毫米波雷达点云的行人重识别技术, 因输入信号固有的稀疏性和信息量的物理限制, 其性能提升受到了极大限制. 为打破这一性能瓶颈, 提出一种新的“模型增强”训练范式: 利用信息更丰富的视频模态作为离线外部知识源, 通过知识蒸馏来提升毫米波雷达行人重识别的精度. 具体地, 设计对称时序建模架构以对齐师生模型的“认知范式”, 并在此基础上提出双重知识引导机制. 一方面, 基于“不同模态下, 人体步态的运动学特征在时序演变上具有一致性”这一关键洞察, 引入时序注意力蒸馏, 显式地指导雷达模型聚焦于步态关键帧. 另一方面, 通过对齐高层决策逻辑传递类别判别信息. 此外, 引入基于实例难度的自适应学习策略, 实现对困难样本的重点辅导. 在自建数据集上的实验结果表明, 所提方法能显著提升雷达模型的识别性能. 相较于最先进的单模态毫米波雷达模型, 在mAP (mean average precision)指标上取得了3.8%的提升.
cs.*
软件学报 · 2026 · 正式发表
张?, 张会彬, 熊晨伊, 王帅, 王怡然, 李志飞
多模态推荐通过联合建模用户历史行为与物品的图像、文本等内容信息, 能够在协同信号之外提供更丰富的语义线索, 从而提升推荐系统对用户偏好的刻画能力. 尽管现有工作已在多兴趣建模、结构学习和跨模态对齐等方面取得了一定进展, 但仍面临若干挑战: 一方面, 部分方法倾向于将用户行为信号与多模态内容压缩到共享表示空间, 导致不同模态下偏好表达的差异性刻画不足; 另一方面, 静态用户-物品交互图在稀疏或噪声较强的场景下容易引入无关邻居传播, 限制表示质量; 此外, 现有对比学习目标多聚焦于同一实体的跨视图一致性或模态内部约束, 对真实用户-物品交互关系的显式利用仍不充分. 针对上述问题, 提出一种融合动态图注意力与多视角对齐的多模态推荐方法DAMARec. 该方法在用户侧采用文本-视觉双通道表示, 以增强差异化兴趣建模; 在结构建模阶段, 引入动态图注意力机制对用户-物品交互关系进行自适应加权, 以突出关键信息传播并抑制噪声干扰; 在此基础上, 从用户侧与物品侧共同构建交互驱动的多视角跨模态对齐目标, 实现协同信号与多模态语义的一致约束. 在3个公开数据集上的实验结果表明, DAMARec在Recall@10和NDCG@10指标上平均提升6.31%和8.18%, 显著优于代表性基线方法, 验证了其在提升表示鲁棒性、增强跨模态语义一致性以及缓解稀疏场景噪声传播方面的有效性.
cs.* · cs.CV
软件学报 · 2026 · 正式发表
谢志文, 张文婷, 付杨杰, 周光有
认知诊断旨在通过建模学生作答行为推断其知识点掌握情况, 对提升个性化教育和学习干预具有重要意义. 现有方法主要依赖统计模型或深度序列模型刻画学生知识掌握变化, 但通常忽视了作答过程中的非认知因素对学习状态的影响, 以及对作答结果的作用. 为解决上述问题, 提出基于学习状态和认知水平融合的双路径认知诊断模型. 在该模型中, 状态路径通过学生非认知行为特征推断学习状态, 并引入监督式对比学习增强状态表征的判别性; 认知路径结合因果分析刻画知识掌握水平与作答结果之间的因果关系; 两条路径通过门控式融合机制实现互补与动态调节, 从而全面反映学生的真实掌握水平. 为了验证所提模型的有效性, 在3个真实教育数据集上进行实验, 结果表明该模型在各类关键指标上均优于9种基线模型, 同时消融实验验证了双路径设计和门控融合在提升认知诊断性能与可解释性方面的有效性.
cs.*
软件学报 · 2026 · 正式发表
廖健, 赵轩, 王素格, 郑建兴, 李德玉
反事实陈述通过假设性和否定性的可能情境传达情绪, 为隐式情绪分析(implicit emotion analysis, IEA)带来独特且复杂的挑战. 当前研究面临3大主要障碍: 缺乏高质量的反事实隐式情绪分析(counterfactual IEA, CIEA)多模态语料库、对未发生场景的语义建模存在困难、解构反事实关联隐喻信息的复杂性. 针对上述不足, 提出CIEA这一新任务, 并构建一个基于多智能体协作推理的多模态语义对齐框架(multimodal semantic alignment framework with multi-agent collaborative reasoning, MOSAMAC). 该框架首先利用大语言模型(large language model, LLM)和多模态大语言模型(multimodal LLM, MLLM)生成视觉场景, 以解构文本中隐含的隐喻信息. 随后, 采用一种词元-图块级的融合方法, 将视觉语义与文本表示进行跨模态细粒度对齐. 此外, 通过设计多智能体协作机制对假设条件进行语义层面的推理, 并将其与多模态特征融合以实现最终的情绪预测. 构建两个新的CIEA数据集, 涵盖中英文并配套了多模态语义资源. 通过与多个具有竞争力的基线模型进行对比实验验证所提方法的有效性.
cs.* · cs.CL · cs.AI · cs.MA · cs.CV
软件学报 · 2026 · 正式发表
Emre Kutlusan, Yongmin Li
Foundation models pretrained on large natural-image collections are increasingly adapted for medical image segmentation, in the expectation that they will replace models trained for each task. Most comparisons, however, use data from the training distribution and a single task-specific baseline whose configuration is rarely justified, and therefore provide limited evidence of performance at new clinical sites. It also remains unclear how much of a foundation model must be adapted, and what determines this. We adapted a foundation model, DINOv2, with strategies ranging from tuning only its bias terms to full fine-tuning, and compared it with U-Nets and ResU-Nets trained from scratch. Experiments covered colonoscopy, breast ultrasound and dermoscopy, with testing on held-out data from the training source and on independent datasets from other institutions. The adapted foundation model generalised better on every external dataset. Within the training distribution, the task-specific models remained competitive when training data were abundant, and delineated boundaries more precisely, while the foundation model more often missed small lesions. Only a small fraction of the foundation model required adaptation, and the capacity needed was determined by the imaging domain rather than by the adaptation method. A widely used polyp test set was also found to duplicate images from another benchmark. Foundation-model adaptation is therefore warranted where models must work across sites, with capacity selected on the target domain.
cs.CV · cs.AI
Hunar Batra, Lachin Naghashyar, Ashkan Khakzar, Philip Torr, Christian Schroeder de Witt, Constantin Venhoff, Ronald Clark
Multimodal Large Language Models (MLLMs) exhibit strong visual understanding, yet the internal features that cause these behaviors remain difficult to identify, audit, or control. While applicable to post-hoc inspection, hidden states that are decomposed into interpretable feature directions using sparse autoencoders (SAEs) neither readily isolate which features are changed by multimodal training, nor are they directly useful for targeted control. We introduce MMDiff, a multimodal model-diffing framework that trains multimodal SAEs and turns them into feature-level interfaces for discovering and controlling multimodal behavior. MMDiff supports three uses: (i) feature isolation, by diffing a base-LM SAE against its multimodal-adapted counterpart to identify features altered by multimodal training; (ii) task-specific feature detection, via per-token contrastive firing analysis that isolates causal features; and (iii) feature-level control, by causally removing or steering the discovered feature directions. We train multimodal SAEs for three MLLM families, LLaVA-MORE, PaliGemma 2, and InternVL3.5, and evaluate on visual-spatial understanding, multimodal safety, and OCR. MMDiff discovers sparse, causally specific features whose removal selectively degrades target behaviors by an average of 12% on spatial tasks and 17% on OCR, and reduces attack success rate by 24% on multimodal safety attacks, with no impact on VQA performance. Steering these features improves spatial and OCR accuracy by +3.6% and +1.8% on average over a standard single-layer steering baseline. These results show that multimodal SAEs can serve not only as interpretability tools, but as mechanisms for auditing, steering, and controlling MLLMs behavior toward safer and more capable generations.
cs.CV · cs.AI · cs.CL · cs.LG
Sheethal Bhat, Bogdan Georgescu, Awais Mansoor, Florin C. Ghesu, Sasa Grbic, Andreas Maier
DeepLearning (DL)-based Computer-Aided Diagnostic (CAD) systems have shown great promise in supporting clinicians, but their performance heavily depends on the availability of large, well-annotated datasets. In medical lesion detection, the scarcity of comprehensive, stan dardized annotations and the heterogeneity of existing datasets limit the development of robust foundation models. In this work, we introduce a novel approach to building a multi-domain foundation model for le sion detection by integrating diverse medical image datasets annotated for both anatomical structures and abnormalities. We first propose a student-teacher training strategy to effectively combine datasets with heterogeneous label spaces, mitigating catastrophic forgetting, and im proving feature learning. In addition, building on our previously pro posed example-based EM-DETR framework, we adapt it to jointly learn from multiple domain-specific datasets, enabling fine-grained, example driven detection across modalities. Our model achieves state-of-the-art (SOTA) results, with image-level AUC scores exceeding previous SOTA by 1% point on four key findings in CXRs. We achieve a leading mAP50 in CXRs and mammography, surpassing EM-DETR trained on a single dataset by 2-3% points. This approach paves the way for scalable foundation models in medical imaging that leverage heterogeneous data while maintaining robust and generalizable performance.
cs.CV · cs.AI
Zhuotao Jin, Xiaoyun Wang, Nicholas Brawand, Roman Zubatyuk, Atul Thakur, Eric Qu, Boris Kozinsky, Justin Smith
The search for new crystalline materials spans an enormous compositional and structural space. Generating candidates in this space requires modeling the relationships between elemental composition, crystallographic symmetry, and crystal geometry. The space group specifies the crystal's symmetry and determines the admissible Wyckoff positions and geometric degrees of freedom; fixing it in advance prevents this global symmetry choice from being revised using information from evolving Wyckoff occupations and elemental composition. We introduce DynaCrys, the first framework for complete crystal generation in which the space group co-evolves with Wyckoff occupations and elements. We develop physics-informed space-group diffusion that explicitly incorporates crystallographic group-subgroup relations, and a legality-constrained stochastic decoder that jointly samples compatible space groups, Wyckoff occupations, and elements, followed by symmetry-constrained geometry generation. In large-scale evaluations on standard benchmarks, DynaCrys achieves state-of-the-art performance in symmetry-aware crystal generation while supporting fast sampling.
cond-mat.mtrl-sci · cs.LG
Alex Shtoff
We present TorchCurves, a PyTorch library that treats vector-valued parametric curves as learnable primitives. Its central operation maps batches of scalar inputs to points on curves represented in B-spline or Legendre polynomial bases, with differentiation through both inputs and coefficients. Curve coefficients can be optimized jointly with the surrounding model or produced by another network. The library provides trainable modules, reusable basis modules, and a functional interface, supporting applications ranging from continuous feature embeddings to Kolmogorov-Arnold networks and functional regression. Its implementation combines classical evaluation algorithms with vectorized PyTorch operations, emphasizing readability and maintainability without custom hardware kernels. Timing experiments compare curve evaluation against NumPy and SciPy and KAN training against EfficientKAN across a range of problem sizes. The results illustrate where this generality remains computationally practical and where overhead is substantial. Library is available on GitHub at https://github.com/alexshtf/torchcurves.
cs.LG · stat.ML
Zijian Zhang, Rizhen Hu, Jiangweizhi Peng, Shiyang Li, Sujay Bhatt, Sihan Zeng, Mingyi Hong
Reinforcement learning with verifiable rewards (RLVR) produces rich training experience, yet outcome rewards reduce each attempt to a single number and leave most of this information unused. Training-time scaffolds, such as strategies or hints added to the prompt during training, offer a natural way to turn this information into guidance, but their design and adaptation are largely left to humans and tailored to specific domains. We introduce AutoScaffold, a unified AI-for-AI framework that turns a student's training experience into adaptive scaffolds, with minimal domain-specific information and no human intervention during training. A frozen teacher periodically examines this experience, diagnoses bottlenecks, and designs scaffolds in two forms: (i) strategies shared across a class of tasks and (ii) hints for individual hard tasks, adjusting the content and amount of support as the student learns. The scaffolds are removed at evaluation. We apply the same framework to ALFWorld, search-based QA, mathematical reasoning, and GPU kernel generation. When evaluated without scaffolds, students trained with AutoScaffold outperform those trained with the corresponding RL baselines in all four domains, with gains of 7.2 points in ALFWorld success and 4.0 points in average QA exact match. On GPU kernel generation, AutoScaffold increases the fraction of tasks with a correct kernel that runs faster than the PyTorch reference by up to 5.0 points. On math, our student surpasses QuestA's reported accuracy using approximately one-third as many student training steps. Analyses show that the teacher identifies a different failure in each task class and shifts its scaffolds from step-by-step workflows to targeted fixes as the student improves, and that its revised scaffolds help the student learn more than keeping the previous scaffold or using no scaffold.
cs.LG · cs.CL
Ziang Liu, James J. Kim, Yijia Dai, Jennifer J. Sun
LLM agents are increasingly used for open-ended, data-driven scientific discovery, yet most of them evaluate hypotheses pointwise, testing an empirical claim $h$ only against its complement $\neg h$. When evidence rejects $h$, the agent learns that one explanation is unlikely but not which alternative the evidence supports. To recover that alternative, we introduce Contrastive Discovery, which replaces the single hypothesis with a contrast set, a scientific question paired with mutually exclusive and collectively exhaustive competing explanations. Each experiment is scored by how its evidence redistributes belief across the set, and the question is resolved when belief concentrates on one explanation. Because contrast sets change only the unit an agent proposes and scores, an existing agent can use them to re-evaluate evidence it has already collected or to guide its search. Across six domains, $89\%$ of the pointwise candidates with the largest belief shifts are rejections. On the same evidence, contrast sets resolve $73\%$ of these candidates to a specific alternative. When contrast sets guide search, the agent resolves $67\%$ more questions than pointwise surprisal under the same experiment budget. In a blinded study, $38$ human experts chose the same explanation on $78\%$ of resolved questions and on $82\%$ of questions where the pointwise agent rejected its hypothesis and stopped. Together, these results support explicit competing explanations as a way for discovery agents to resolve more scientific questions from the same evidence and validation budget.
cs.AI · cs.CL · cs.LG
Ken-ichi Kawarabayashi, Hirotaka Yoneda, Masataka Yoneda
We study online coloring, focusing on the competitive ratio. The best previously known results were (1) a lower bound of $Ω(n / \log^2 n)$ for both deterministic and randomized algorithms by Halldórsson and Szegedy (1994), (2) an upper bound of $O(n / \log n)$ for randomized algorithms by Halldórsson (1997), and (3) an upper bound of $O(n / \log \log n)$ for deterministic algorithms by Kawarabayashi, Yoneda, and Yoneda (2025). In this paper, we improve these results as follows: 1. A new lower bound of $Ω(n / \log n)$ for deterministic algorithms. 2. A new upper bound of $O(n (\log \log n)^2 / \log^2 n)$ for randomized algorithms. 3. A new upper bound of $O(n / (\log \log n)^{2-\varepsilon})$ for deterministic algorithms for every fixed $\varepsilon>0$. For randomized algorithms, we nearly close the gap between the lower and upper bounds; the remaining gap is only $O((\log \log n)^2)$. We establish the asymptotic separation between randomized and deterministic online graph coloring. Against an oblivious adversary, a polynomial-time randomized algorithm achieves a competitive ratio asymptotically smaller than any deterministic algorithm can guarantee, even with unlimited computation.
cs.DS · cs.DM · math.CO
Xichen Pan, Aashu Singh, Satya Narayan Shukla, Xiangjun Fan, Shlok Kumar Mishra, Saining Xie
Large language models (LLMs) are widely used in text-to-image (T2I) systems, but they are typically limited to text encoding, while denoising is handled by newly trained generative backbones. The emergence of representation autoencoders (RAEs) shifts the generation target toward semantically structured visual representations, creating a latent space that is more compatible with pretrained LLM priors. Inspired by multimodal LLMs (MLLMs), where an MLP projector is sufficient to align clean visual representations with a pretrained LLM, we repurpose the MLLM itself as a noisy representation encoder, extending this mechanism from clean to noisy inputs. We present RepFusion, which uses the resulting MLLM outputs as the conditioning signal for a diffusion transformer. In controlled comparisons at similar inference budgets, RepFusion outperforms baselines that devote comparable capacity to newly initialized denoisers. Reusing MLLM outputs across denoising steps further accelerates inference while retaining performance comparable to per-step MLLM recomputation. These results demonstrate that MLLMs provide strong priors for denoising visual representations.
cs.CV
Samuel German
The Path Avoiding Forbidden Pairs problem (PAFP) asks whether, in a directed graph $G$ with two of its vertices labeled $s,t$ and a set $\mathcal F$ of forbidden vertex pairs, there is an $s$--$t$ path that contains at most one endpoint from each pair in $\mathcal F$. We initiate the study of PAFP through a layer-based width measure. We first look at the directed union digraph $G\cup\mathcal F$, which is obtained by adding one arc per forbidden pair to $G$, oriented according to a fixed reachability-compatible order. Let the BFS layer $L_d$ be the set of vertices at directed shortest-path distance $d$ from $s$, and define the BFS-width from $s$ as $\max_d |L_d|$. We show that if $G\cup\mathcal F$ has BFS-width $b$ from $s$, and $p$ is the maximum, over BFS-layer cuts, of the smaller of the numbers of distinct heads and distinct tails of backward arcs crossing the cut, then PAFP is FPT parameterized by $b+p$. Here a backward arc goes from a later BFS layer to an earlier one. In contrast, we show that bounded BFS-width alone does not suffice: PAFP remains $\mathsf{NP}$-complete when the union digraph has BFS-width 2. We also study layer-based width parameters on the input graph. We show that if the input DAG has BFS-width at most $2$ and only $k$ backward input arcs, then PAFP can be decided in $2^k |I|^{O(1)}$ time, with unrestricted forbidden pairs, and this result is tight. Moreover, we study exact-length layers in the input graph, where the $d$-th exact-length layer consists of the vertices reachable from $s$ by a directed path of length exactly $d$. We show that for DAGs of exact-length width at most $2$, PAFP is polynomial-time decidable by a $2$-SAT encoding of fixed-length paths. Unlike previously known polynomial-time regimes our 2 input -graph tractability results simultaneously allow unrestricted forbidden pairs and input graphs with exponentially many $s$--$t$ paths.
cs.DS · cs.DM
Megha Srivastava, Michael H. Lim, Allison M. Keys, Pranav Murugan, Nate Gruver
Predicting how a ligand binds to its target is central to small-molecule drug discovery. Modern cofolding models can generate diverse candidate poses, but reliably ranking these candidates remains challenging. We introduce Pyrite, a pairwise pose-ranking benchmark comprising 1,500 pose pairs across diverse protein-ligand complexes. Expert chemists equipped with 3D visualization tools and physics-based scores perform near chance, highlighting the difficulty of interpreting available chemical signals. To investigate automated ranking, we introduce PyriteRep, a structured representation of molecular geometry, protein-ligand interactions, and physicochemical scores, and develop two LLM-based agents: PyriteReasoner, which directly compares poses through text-based reasoning, and PyriteFeaturizer, which generates executable code to construct features for logistic regression, optionally incorporating features derived from LLM subcalls. These agents achieve up to 67% and 74% accuracy, respectively, exceeding observed expert performance. LLM-guided feature discovery generally outperforms direct reasoning, while Group Relative Policy Optimization (GRPO) improves open-weight models, enabling them to match or surpass closed-source frontier models in our evaluations. These results establish Pyrite as a challenging test of pose-ranking ability and highlight LLM-guided feature discovery as a promising approach to interpret and combine the outputs of computational chemistry tools.
cs.LG · q-bio.QM
Yang Cai, Yingkai Li, Jinzhao Wu
We study the allocation of indivisible items to buyers with additive values drawn from independent distributions. The objective is consumer surplus net of burnt payments. We consider artificial-currency mechanisms: buyers receive personalized budgets of a currency worthless outside the mechanism and use them to bid across items. The better of separate selling and artificial-currency mechanisms provides a 3-approximation to optimal consumer surplus. The guarantee holds at every Bayes--Nash equilibrium, without equilibrium selection. Both branches are necessary: either alone may capture only a vanishing share of the optimum. The proof decomposes benchmark surplus into a stable part captured by artificial currency and a variation covered by separate selling.
cs.GT
Yeyuan Liu
Connectome-constrained models are often used to argue that a measured wiring diagram outperforms shuffled or rewired wiring, and hence that structure shapes function. Such comparisons can confound wiring with how the models are tuned: a connectome fixes who connects to whom, not the excitability, gains or plasticity rates the model also needs, and modellers often tune these until the real-wiring model works. We separate three estimands: physiology tuned on the real wiring and transferred to nulls (P1), each wiring tuned on its own (P2), and P2 restricted to physiology consistent with independent data (P3). The biases involved are known in machine learning (under-tuned baselines, inherited weights, winner's curse); we measure what they do to biological conclusions, using synthetic ground truth, placebo nulls and three Drosophila circuit models. Some claims survive every protocol we test: segregated reward and punishment dopamine improves learning, and head-direction topography is required for a persistent bump. Others do not. In the mushroom body, the real Kenyon-cell output wiring beats a cross-subtype identity shuffle by 0.09 under P1 and P2 but loses to it by 0.09 under P3 (95% CIs exclude zero; held-out runs confirm). The verdict is set by the share of alpha'beta' cells among active Kenyon cells, which P1 selects low; at the share implied by imaging (about 44%) the real-wiring model fails to learn, exposing a model inadequacy rather than a deficit of the fly's wiring. In flyvis, P1 penalises a mirrored connectome, which has a constructed solution within 0.07 (median) of the real error, more than a spatially shuffled one in 50 of 50 members; the ranking survives ten-fold longer retraining but depends on the checkpoint at fifty-fold. A credible structural claim needs matched optimisation, data-constrained physiology and a model that still works there.
cs.LG · q-bio.NC
Daniel Lenton
Agent harnesses have emerged as a promising mechanism for continual learning, alongside test-time parametric learning. However, the current benchmarks have several limitations: they generally focus on a single applied domain, serve each instance only once, and assume a single family of learner, so memory-based and weight-based systems are rarely compared on the same scale. We propose Continual-ARC, a lifelong stream built from the three ARC-AGI editions, whose tasks need no domain knowledge. The same tasks return after dormant gaps with fresh instances, while new tasks keep arriving. On each instance, the learner may pay for demonstrations, submit answers, or act inside the harness, with retries permitted on incorrect submissions, which are paid for. Each instance is scored based on the share of this feedback budget consumed. Since a learner may keep anything between instances, memory-based and weight-based systems are scored on the same scale. After the stream, the learner is frozen and tested on fresh instances of the tasks it met, which measures memorization, and on unseen tasks, which measures generalization. We run an agent harness with frozen weights, with and without persistence, on Qwen3.5-9B and GLM-5.3 Flash, comparing them with two prompting floors, a program-induction method and Qwen3.5-9B's weights trained to imitate GLM-5.3 Flash. Persistence in the harness generally does not improve memorization unless provided with explicit expert guidance. Most harness-based methods still learn less than a floor that simply caches each task's demonstrations, across all models we tested. No harness-based method gains on unseen tasks, whereas weight-based learning shows substantial gains. This opens the door to future weight-based continual learning that could teach models to drive harnesses in new domains.
cs.LG
Margalit Glasgow, Joan Bruna
We consider one-hidden layer neural networks trained in the feature-learning regime using gradient descent, and relate the output of the finite-width network $f_{\hatρ_t^m}$ to its infinite-width counterpart $f_{ρ_t^{MF}}$, which evolves in the mean-field dynamics. While constant-time horizon bounds for $\|f_{ρ_t^{MF}} - f_{\hatρ_t^m}\|$ may be obtained via standard Grönwall estimates, the long-time behavior of the fluctuation is a more delicate matter. Uniform-in-time bounds often rely on (local) strong convexity in the landscape or Logarithmic Sobolev inequalities present in noisy gradient dynamics. In this work, we establish non-asymptotic weak propagation-of-chaos that holds uniformly in time, obtained by exploiting instead the convergence rate of the mean-field deterministic Wasserstein-gradient-flow dynamics. Specifically, denoting by $L_t$ the mean-field excess MSE loss at time $t$ and $m$ the number of neurons, under standard regularity assumptions and the condition $\int_0^\infty L_t^{1/2} dt =O(\log d)$, we obtain the uniform in time bound $\|f_{ρ_t^{MF}}- f_{\hatρ_t^m}\|^2 \lesssim \text{poly}(d) m^{-\min(1,c/6)}$ whenever $L_t \lesssim t^{-c}$. Our result holds in a noiseless setting and does not make any assumptions on the geometry of the landscape near the optimum, and extends seamlessly to other forms of discretization, including finite number of samples and time discretization. A key takeaway of our result is that whenever the convergence rate of the mean-field, population-loss dynamics is faster than $t^{-2}$, we can attain a loss of $ε$ with only $\text{poly}(d/ε)$ neurons, training samples, and GD steps.
stat.ML · cs.LG
Adi Mashiach
Foundation models trained on large unlabeled corpora develop emergent capabilities, tasks that go beyond the training objective. We extend this paradigm to multimodal biology with Bonbon, a foundation model for molecular interactions trained on approximately 3.2 trillion tokens of paired protein-ligand sequences. From a single representation space, Bonbon exhibits three zero-shot emergent capabilities at hierarchical levels of resolution: functional (mechanism of action), residue (binding site localization, including orthosteric/allosteric discrimination), and bond (covalent versus non-covalent engagement). All three capabilities generalize to out-of-distribution complexes. Bonbon was trained on protein-ligand sequences alone, without any task-specific supervision and without structural coordinates. We conducted controlled scaling experiments across six model scales from 42M to 1.6B parameters; every capability evaluated shows a sharp jump at the largest scale on continuous metrics, even as pretraining loss decreases smoothly. This demonstrates that loss-based scaling laws alone are insufficient to predict when zero-shot emergent capabilities become available. The capabilities reported here emerge after training at a 1:2,000 parameter-to-token ratio to accommodate the sparsity of paired biological interaction data. The same frozen representations achieve state-of-the-art binding and affinity prediction at five orders of magnitude greater speed than co-folding methods. In prospective wet-lab validation across nine disease targets spanning seven protein families, Bonbon achieves a 61.5% dose-response hit rate on de novo compounds, yielding structurally novel molecules not previously reported to inhibit these proteins, including for out-of-distribution targets.
cs.LG · q-bio.BM
Yogesh Kumar, Akshay Ankush Yadav, Susanta Samanta
In AES-like ciphers, diffusion layers are commonly instantiated using MDS matrices, since their optimal branch number yields strong diffusion guarantees and underpins classical resistance arguments against differential and linear cryptanalysis. However, Daemen and Rijmen (2009) showed that linear layers may still exhibit related-differential structure beyond what the MDS criterion captures, and Bardeh and Rijmen (2022) demonstrated that this phenomenon can be exploited in attacks on reduced-round AES. In this work, we systematically investigate the conditions under which linear layers avoid or exhibit these differentials, identifying matrix classes for which such structure is unavoidable. We first prove that every non-MDS matrix admits a nontrivial pair of related differentials, showing that the MDS property is necessary for avoiding them. We then establish that every odd-order symmetric MDS matrix admits related differentials, which rules out broad families of Cauchy-based constructions. We also substantially strengthen the circulant case by proving that related differentials are unavoidable for every circulant matrix of order $n$ with $n \not\equiv \pm 2 \pmod{12}$. Finally, we revisit the characterization of $3 \times 3$ MDS matrices over $\mathbb{F}_{2^m}$ for the absence of related differentials, and derive an explicit necessary and sufficient criterion in terms of $15$ polynomial constraints.
cs.CR
Sandro Martinelli Reia, Taylor Anderson, Dieter Pfoser
Characterizing how people move through a city requires spatially and temporally fine-grained data that are often costly and difficult to obtain. Here, we introduce a computational framework that uses macroscopic, aggregated, and static urban data to construct a microscopic, dynamic, and spatially detailed representation of urban mobility. Building on an existing activity simulator, the framework combines a synthetic population, the built environment, and transport networks to generate daily trips and assign walking, car, or public transit. Applied to approximately 475,000 synthetic residents of Niterói, Brazil, simulated neighborhood-level transit patterns show a strong association with an empirical activity proxy (Pearson r=0.885). Walking is more prevalent in compact central areas, car use increases toward the periphery, and public-transit use remains comparatively high along bus corridors. Reducing walking times to and from bus stops by 30% increases the modeled public-transit-to-car ratio by 11.3%, with larger transit-share gains in several peripheral areas. These results reveal spatial differences in transportation choice and identify where improved transit access may have the greatest modeled effect.
physics.soc-ph · cs.MA · stat.OT
Alexandre Garcia-Duran, Alejandro Rodriguez-Garcia, Manuel Molano-Mazón, Alexandre Hyafil, Srikanth Ramaswamy, Janne K. Lappalainen
Biological visual systems compute reliably under tight metabolic and wiring constraints, whereas artificial vision systems consume far more energy and can be sensitive to adversarial perturbations. Whether networks with brain-like wiring are more robust at lower computational cost is unclear. We address this by benchmarking a task-optimized, continuous-time recurrent neural network of the fly visual system constrained by its synaptic connectome (FlyNet) against artificial neural networks. We compare it against CNNs, RAFT, and continuous-time recurrent neural networks without detailed connectome constraints (sCTRNN, CTRNN) on a collision-detection task within a differentiable 3D rendering pipeline. FlyNet is adversarially vulnerable but, relative to its compute, more robust to common video corruptions than RAFT, while using up to five orders of magnitude fewer FLOPs. The CTRNNs remain more robust than FlyNet across almost all tested perturbations, suggesting that connectome constraints benefit efficiency, through their low edge count, over robustness. Recurrence improves robustness to white-box attacks in RAFT and temporal context improves robustness to common corruptions. FlyNet's position and velocity representations degrade under adversarial textures but recover under counterfactual textures that restore correct classification while staying equally far from training, ruling out unfamiliar textures as the cause of failure. The reverse-phi illusion similarly disrupts these representations and inverts the same low-level motion statistic as the adversarial textures. Optimizing this input statistic alone, without access to the network, suffices to induce classification failure, attributing both perturbations to the same circuit vulnerability. Connectome-constrained models can thus surface adversarial failure mechanisms as testable hypotheses about biological perceptual vulnerabilities.
cs.CV · q-bio.NC
Tanishq Rachamalla, Srishti Kaushik, Muhammad Attique Khan, Gustau Camps-Valls, Swalpa Kumar Roy
Cross-sensor remote sensing image retrieval aims to find scenes with the same content across heterogeneous sensors, such as SAR and optical or panchromatic (PAN) and multispectral (MS) imagery, without manual labels. Existing self-supervised methods either reconstruct sensor-specific pixel detail, align each image only with its co-registered twin, or rely on large transformers with 118-173M parameters (and up to 398M). We introduce RS-JEPA, a compact joint-embedding predictive framework built on a Vision Mamba encoder shared by both sensors. Since a state-space encoder reads an image as an ordered sequence of patches, we propose scan-path views, which treat two reading orders of the same image as two views with identical pixels and predict the hidden patches in one order from the visible patches in the other.To keep this prediction challenging, we propose Adversarial View Coupling (AVC), which learns the hardest split of the visible patches, and a neighbourhood consensus loss that aligns the scene similarities of both sensors. Extensive experiments on SAR-optical (BEN-14K, CBRSIR-VS) and PAN-MS (DSRSID) benchmarks demonstrate that RS-JEPA achieves the highest F1@5 among twelve re-implemented methods (72.92, 80.35 and 94.43) with a retrieval network of only 3.88M parameters (14.57M in training, 8-16times fewer than prior JEPA retrievers). Like prior methods, RS-JEPA trains on co-registered pairs. Our results highlight reading order as a source of self-supervised views. code: https://github.com/Tanishq251/RSJEPA.
cs.CV
Aditya Saboo, Pratibha Verma
Skipping zero activations saves multiplications in CNN accelerators, but it complicates accumulation: the products in a compressed input window belong to different output columns and must be merged. In our earlier engine, this merging was done by read-modify-write (RMW) of a row buffer, which made sparse inputs slower than dense ones even at 87% sparsity. Existing designs either merge kernel rows through RMW or reduce bank conflicts with extra hardware, without eliminating them. SkipLane makes the entire accumulation path write-once. Each kernel row is processed in its own lane, which holds partial sums in three registers and retires up to three finished columns per cycle into memory banks that cannot conflict. Lanes and input channels are summed only at readout, so no partial sum is ever read back. Multipliers left idle by zero weights compute a second filter in the same pass. In RTL simulation, the engine stays within 0.5% of its frontend bound up to 70% sparsity and is 1.10-1.26$\times$ faster than an SCNN-style scatter accumulator with 18 banks and ideal RMW at 50-80% sparsity. Six engines meet 140 MHz on a PYNQ-Z2 using 21.1k LUTs and run LeNet-5 bit-exact on the board. For conv1, the engine needs 1.94$\times$ fewer cycles than a dense engine (1.26$\times$ faster end to end); for conv2, whose rows are only 13 pixels wide, it reaches 0.87$\times$ the dense speed.
cs.AR
Howard Zhang, Chuck Arvin, Logan Brown, Mihir Bhatia, Neelu Choudhary, Lily Miao
Deploying LLM-based analytics agents in enterprise settings requires evaluation frameworks that can reliably detect failures across complex, multi-tool workflows. We present a three-phase comparative study of three evaluation frameworks, each representing a distinct evaluation paradigm (trace-based LLM judging, text-only LLM judging with red-teaming, and deterministic heuristics), applied to two analytics agents in a controlled research setting using frozen execution traces. Phase 1 quantifies evaluation harness stability through repeated measurements, finding that 97% of configurations meet the stability threshold and that stability patterns are agent-independent. Phase 2 measures known risk detection on 72 ground-truth test cases, where the trace-based framework achieves the best precision-recall balance (F1=0.90) while text-only evaluation matches on recall but generates 5x more false alarms. Phase 3 evaluates unknown risk discovery, uncovering 24 novel failures with zero overlap between functional testing and security red-teaming. Our results demonstrate that (1) the information available to the evaluation harness determines evaluation quality, (2) functional and security evaluation are complementary, and (3) LLM-based evaluation harnesses surface subtle errors that human reviewers miss. We recommend a two-framework strategy: trace-based evaluation for functional assessment and adversarial red-teaming for security validation.
cs.AI · cs.CL · cs.LG · cs.SE
Kaiyuan Hou, Yang Ping Kuo, Indranil Bhattacharya, Rohith Nama, Arushi Gupta, Carlos D. Castillo, Prathap Ramachandra, Xiaofan Jiang 等 9 位作者
Business process automation runs the same Standard Operating Procedure (SOP) again and again on new inputs. Large Language Model (LLM) agents can follow an SOP written in natural language, but they call an LLM to choose every step of every run, which adds latency and cost at scale. We present ReTrace, which turns an SOP and a few logged agent runs into a reusable workflow in which code, not an LLM, selects the next step. The agent's reasoning, the explanation it writes with each tool call, often states why it took an action; the tool calls alone do not. ReTrace extracts the condition behind each step, merges these conditions across runs into a compact rule document, and generates an executable workflow from it. Only this last stage depends on the platform, so rule documents can be compiled for different orchestration platforms. On $79$ procedures from our production platform, ReTrace workflows reach $93.8\%$ completion, against $96.8\%$ for the agent, at $24\%$ lower cost per task. On $τ^2$-Bench, they reach $75.8\%$ task success (agent: $80.6\%$) and run in $1.3$s instead of $170$s; three rounds of repair on training failures lift $k{=}2$ workflows from $66.7\%$ to $81.2\%$. On eight SOP-Bench procedures, ReTrace without repair trails few-shot generation ($80.6\%$ vs. $83.6\%$), but repair from labeled training failures lifts it to $87.8\%$ at $0.40$s per task, compared to $85.5\%$ in $46.6$s for the agent.
cs.AI · cs.LG · cs.MA
Peng Chen, Yihang Wang, Zhongwen Rao, Yang Shu, Yuying Qiu, Yuxuan Chen, Lujia Pan, Bin Yang 等 9 位作者
Time series analytics has been widely applied across various domains. While existing time series foundation models (TSFMs) have demonstrated promising results, they still face two major challenges: 1) Most TSFMs mainly rely on stacking Transformer layers to increase parameter size, lacking the ability to adaptively model complex time series patterns from multi-domain data. 2) Current TSFMs are primarily pre-trained for specific tasks, e.g., forecasting or anomaly detection, and a few models support multiple tasks, but they fail to fully leverage task specificity and commonality, resulting in suboptimal performance. To address these challenges, we propose TimeSeek, a novel generalized time series analytics model that adaptively models time series patterns from multi-domain data and unifies multiple tasks in pre-training. Conceptually, TimeSeek is guided by selective information sharing under domain and task heterogeneity. From the perspective of multi-domain data, we propose a Multi-Domain Feature Extractor that adaptively captures temporal dependencies and channel correlations based on domain-specific characteristics, including progressive multi-scale modeling and adaptive slot channel modeling. From the perspective of the diversity of tasks, we propose multi-task collaborative training, which leverages task specificity and task commonality through task customization and cross-task training, effectively unifying multiple tasks in pre-training and better adapting to diverse downstream scenarios. After pre-training on large-scale time series data, TimeSeek achieves state-of-the-art performance in forecasting, anomaly detection, and classification tasks across both few-shot learning and zero-shot situations, together with favorable inference efficiency.
cs.LG
Dirk Bergemann, Yang Cai, Kangning Wang, Jinzhao Wu, Renzhe Xu
We prove that at least half of the first-best gains from trade can always be achieved in multidimensional two-sided markets with additive values and costs and arbitrary downward-closed feasibility constraints. The guarantee is achieved by a mechanism satisfying Bayesian incentive compatibility, interim individual rationality, and strong budget balance, and is tight already in single-item bilateral trade. Types are independent across agents, while the coordinates of each agent's type may be arbitrarily correlated. Feasibility constraints may couple trades across items and agents. This extends the known tight one-half guarantee for single-item bilateral trade. Our proof develops a new application of the Cai--Devanur--Weinberg duality framework. We show that the objective of every feasible dual is at least half of the first best. Strong duality then yields the guarantee without characterizing an optimal mechanism or dual. The key observation is that every feasible dual induces continuous-time Markov chains that, when stopped by exponential clocks, recover true types as expectations of virtual types. The dual constraints control the rate at which these chains change the prior distributions. We establish our bound by combining this control with a coupling of the agents' clocks.
cs.GT
Chubin Zhang, Bo An
Memory-augmented vision-language-action (VLA) policies are meant to remember what happened. Many of them, however, fill memory by time: they sample frames uniformly across the episode and stamp each frame with its time step. Because time passes even when nothing happens, such a memory may not tell waiting from working. We show that it does not. On the counting tasks of RoboMME, we make the robot briefly pause or fidget its wrist. Such a null event changes nothing the task depends on, yet a released memory VLA counts the interval as work done and stops early. We call this failure phantom progress. It cuts success from 47 to 16 of 50 on one task and from 44 to 10 on the other. When the same frames are withheld from memory, success returns, so the cause lies in memory rather than in the interruption itself. We trace the failure to a temporal shortcut: because demonstrations proceed at a steady pace, elapsed time predicts progress. Indeed, rescaling the time stamps alone makes the policy stop early when the clock runs fast and repeat when it runs slow. Surprisingly, removing the time encoding does not cure the failure, because the frames of a null event still occupy slots of the time-sampled memory. Across RoboMME's released designs, vulnerability follows how memory is indexed. Of the fixes the diagnosis suggests, training with injected pauses does not remove the failure and impairs counting, whereas indexing memory by events recovers half or more of the loss without retraining. A memory meant to remember what happened should count events rather than clock ticks.
cs.RO