Machine Learning
Exact Recovery Thresholds for Weighted Data Selection in Vector-Valued Linear Regression
Guangjian Zhang
We resolve the threshold part of Question 4 of the COLT 2025 open problem "Data Selection for Regression Tasks" of Hanneke, Moran, Shlimovich and Yehudayoff. We study vector-valued linear regression with square loss $\ell_{(x,y)}(W)=\lVert Wx-y\rVert_2^2$, where $x\in\mathbb{R}^d$ and $y\in\mathbb{R}^m$. The learner returns the minimum-Frobenius-norm empirical risk minimizer. We prove that the minimal budget of weighted examples for recovering the full-data loss on every finite
dataset is exactly $n^{\star}(d,m)=(m+1)d$. We determine the weighted selection profile $F_{\mathrm{weighted}}(d,m,n)$ at the near-threshold budget: $F_{\mathrm{weighted}}(d,m,(m+1)d-1)=1+1/(dm^2)$. We recover the known spanning-budget value $F_{\mathrm{weighted}}(d,m,d)=d+1$ for every $m$, and $F_{\mathrm{weighted}}(d,m,n)=\infty$ for $n<d$. For the smallest open intermediate
cell $(d,m)=(2,2)$ we prove $F_{\mathrm{weighted}}(2,2,3)\in[13/8,15/8]$ and $F_{\mathrm{weighted}}(2,2,4)\in[5/4,3/2]$. We reduce the conjectured exact values $13/8$ and $5/4$ to a finite moment problem on the circle with at most seven atoms and assemble structural evidence for it. The upper bounds use a fixed-basis conic compression lemma, a determinant–facet
rigidity theorem for maximal certificates, and sharp sparsification lemmas for zero-mean weighted point systems. These tools may be of independent interest. We also exhibit an explicit six-point integer
dataset with $d=m=2$ on which no weighted selection of $2d$ points recovers the optimal loss. Thus the scalar sufficient budget $2d$ does not extend to vector-valued outputs. Our new regression-profile results for $m\ge2$ extend the scalar theory for $m=1$.
Read the full abstract ▾
Numerical Analysis
Mesh-Uniform Power Stability of Two-Relaxation-Time Vector Lattice Boltzmann Schemes with Reversible Boundaries
Jin Zhao
We consider the collision-transport operator for vector-valued two-relaxation-time
lattice Boltzmann schemes linearized about a uniform rest state. Assume that the
equilibrium blocks are positive definite and that the link-even and link-odd relaxation parameters satisfy s_+ + s_- = 2 and 0 < s_- < 2. If the homogeneous transport is
unitary in the
equilibrium metric and reversible under velocity exchange, then the powers of the amplification operator are bounded uniformly with respect to the number and arrangement of
lattice nodes. The admissible transports include periodic transport, vector halfway bounce-back, coordinate-aligned specular reflection, tangential orthogonal involutions, and compatible multi-channel scattering. The proof reduces the population equation to a two-step macroscopic recurrence generated by a contraction. An inclusion of the numerical range in an ellipse, combined with the Crouzeix-Palencia
theorem, yields a dimension-independent estimate for the companion operator and hence the population bound. For a three-coefficient off-midpoint boundary interpolation, we give an exact rational D2N5 example whose finite-domain amplification
matrix has a real
eigenvalue larger than one, although the interpolation coefficients and bulk parameters are admissible. Thus coefficient convexity alone does not ensure stability for this boundary family.
Read the full abstract ▾
Machine Learning
Strong Drafts Need Compact Memories: Long-Context Speculative Decoding with Compressed KV Cache
Tong Yuan, Chengxi Liao and Zeyi Wen
Long-context LLM applications such as document summarization and multi-turn
agents require generation from prefixes spanning tens of thousands of
tokens, making decoding latency a major bottleneck. Speculative decoding (SD) reduces latency without changing model outputs, but its speedup depends on both accepted draft
tokens and draft-step latency: Lightweight drafts are fast but lack the capacity to capture long-range dependencies, whereas strong independent drafts recover acceptance but incur growing KV-access cost at long prefixes. We introduce
memory-augmented drafting for long-context SD, equipping a strong independent draft with compressed draft-side KV
memory: A lightweight adaptor constructs and incrementally updates this
memory to retain distant information and exact recent context. The target verifier retains its full KV cache and applies the standard accept/reject rule, preserving SD's lossless guarantee. Experiments on Llama~3.1-8B and 70B targets at prefix lengths up to 32K show that our method reduces draft-side
memory by over 70%. It achieves speedups of up to 2.08x and 3.33x , respectively, over autoregressive decoding.
Read the full abstract ▾
Information Retrieval
SetMIR: Multi-Interest Retrieval as Set Prediction
Xiaodong Liu, Congfei Zhang, Hsiang-wei Chao et al. (+13 more)
Embedding-based
retrieval is at the core of industrial recommender systems, but a single user
embedding is often too limited to capture a user's diverse interests. Multi-interest
retrieval addresses this by using multiple user
embeddings, yet existing methods still suffer from two issues: interest collapse, where different
embeddings learn the same interest, and static dispatch, where serving uses a fixed
retrieval budget even when some
embeddings are unnecessary. We propose SetMIR, which treats multi-interest
retrieval as a set prediction problem. SetMIR encodes a user's behavior history with a
transformer and uses K learnable queries to decode a set of user interests, each producing a
retrieval embedding and a presence score. During training, Hungarian matching assigns targets to queries one-to-one, so matched queries learn distinct interests and the presence head learns which queries are active. At serving time, SetMIR uses presence scores and query-level Non-Maximum Suppression (NMS) to issue only active, non-redundant ANN queries. On Snap's Dynamic Product Ads (DPA) data, SetMIR outperforms four learned multi-interest retrievers on every metric while issuing 33% fewer ANN queries per request. Deployed as a new
retrieval source in the DPA production stack, SetMIR lifts overall CVR by 3.1%, while lifting CTR by 44% and CVR by 51% over the item-to-item
retrieval source with the same item
embeddings, ANN index, and
retrieval quota.
Read the full abstract ▾
All authors Xiaodong Liu, Congfei Zhang, Hsiang-wei Chao, Siman Wang, Xiao Bai, Tong Zhao, Jingxiao Ma, Wen Zhang, Zhe Liu, Shantanu Aggarwal, Di Huang, William Leach, Yunzhi Zhou, Yajun Wang, Jinchao Li, Yu Zhang
Artificial Intelligence
Generating Workflow DAGs from Natural Language with Non-Reasoning LLMs
Anand Iyer, Bhanu Khetharpal, Srinivas Upadhya et al. (+1 more)
This paper addresses the problem of translating natural-
language routing rules written by business administrators into executable workflow graphs for enterprise contact centers. Each target is a directed acyclic graph (DAG) of conditional actions with parallel branches, hit-first fallback chains, and per-branch Boolean predicates, encoded in the JSON dialect of a commercial routing platform. We show that neuro-symbolic decomposition enables lower-cost, non-
reasoning large
language models to generate complex workflow DAGs at production-relevant quality without expensive extended-
reasoning models. Our central diagnostic is an emission-density bottleneck: on a 635-rule
benchmark of manufactured synthetic data, models select the correct graph nodes with high accuracy but increasingly misconfigure attributes and Boolean grouping as the number of interdependent nodes emitted in one pass grows. We therefore move combinatorial graph construction from the model into a deterministic compiler driven by a compact intermediate representation, with a learned registry-selection front end that focuses generation on relevant vocabulary. Across four models, the full system reaches approximately 89% LLM-judge validity, approximately 90% exact-match condition accuracy, and 99-100% valid JSON while using roughly half the per-rule
prompt tokens of a monolithic
prompt. On GPT-5.3-chat, the method improves judge validity by 24 percentage points and achieves statistical equivalence to a
reasoning model's out-of-the-box quality, although an approximately 8-point frontier gap remains. We also present a deployment path and transferable lessons for structured-generation applications.
Read the full abstract ▾
All authors Anand Iyer, Bhanu Khetharpal, Srinivas Upadhya, Ramkumar Rajagopal
Software Engineering
DSEffi-Bench: Demystifying Large Language Models' Capability in Efficient Data Science Code Generation
Zhihao Gong, Junzhe Yu, Dong Huang et al. (+3 more)
Current data science (DS) code generation
benchmarks equate correctness with quality, overlooking execution time differences that span orders of magnitude between correct solutions. We introduce DSEffi-Bench, the first
benchmark specifically targeting execution efficiency in LLM-generated DS code, comprising 1,000 instances across 10+ DS libraries with stress-testing harnesses and human-validated references. Evaluating 16 models across 3 tiers, we find that correctness alone fails to characterize efficiency: GPT-5.4 leads in correctness (Pass, 66.9%) but its efficiency score (B$|$P, 71.7%) nearly matches GPT-5.4-mini (71.6%), which solves 47 fewer tasks; Kimi-K2.5 ranks lowest in correctness among frontier models (40.2%) yet achieves the highest efficiency score (73.6%) across all 16 models. A human-annotated five-category taxonomy reveals that 79.1% of efficiency deficits extend beyond algorithmic complexity to domain-specific root causes, with distinct failure profiles across model tiers and libraries. Two exploratory experiments provide initial evidence that these diagnostics can guide improvement, yielding up to +14.7% efficiency gains via taxonomy-guided optimization and approaching Claude-Opus-4.6 Best@3 in efficiency at 13.0$\times$ lower cost via library-conditioned routing.
Read the full abstract ▾
All authors Zhihao Gong, Junzhe Yu, Dong Huang, Zeyu Sun, Jie M. Zhang, Dan Hao
Computer Vision
OPUS: A Simple yet Effective Unified Framework for Open-Vocabulary Detection
Xiaoyan Wei, Zhimin Yao, Ruilin Yang et al. (+4 more)
Recent unified open-vocabulary detection (OVD) supports heterogeneous
prompts, including text queries, visual exemplars, and their combinations, but often rely on increasingly complex designs such as heavy cross-modal fusion, staged training, and iterative annotation pipelines. We revisit whether such complexity is necessary in the era of stronger foundation models. Our finding is that unified OVD can be made substantially simpler with semantic-rich visual representations and scalable grounding supervision. We present OPUS (Open-vocabulary,
Prompt-Unified, Simple), a unified detector supporting text, interactive visual, generic visual, and mixed prompting within one framework. OPUS adopts a simple three-part design. Its model architecture combines a semantic-rich visual encoder, built on a DINOv3-ConvNeXt-B backbone with efficient hybrid encoding, with a
prompt-aware decoder that avoids
prompt-specific branches for unified
prompt reasoning. OPUS is trained with a one-stage text-visual training strategy with Instance-level Contrastive
Alignment (ICA), and is supported by a SAM3-based single-pass data engine for heterogeneous grounding supervision. Experiments on COCO, LVIS-minival, and ODinW35 show that OPUS achieves state-of-the-art Visual-I performance, reaching 68.1/69.2/54.7 AP, while maintaining balanced Text and Visual-G accuracy. OPUS also turns mixed prompting from interference into complementarity, improving over text or visual
prompt alone. These results show that simplicity and strong unified prompting capability can be achieved together.
Read the full abstract ▾
All authors Xiaoyan Wei, Zhimin Yao, Ruilin Yang, Wei Zhang, Yong Dai, Yi Zhang, Wei Ge
Machine Learning · Cross-listed from math.LO
A Borel Concept Class of VC Dimension One with a Non-PAC Consistent Learner in ZFC
Mateus Jesus de Arruda Campos, Gabriel Fernandes and Vinicius de Oliveira Rodrigues
The fundamental
theorem of statistical learning states that, under suitable measurability assumptions, finite Vapnik–Chervonenkis (VC) dimension guarantees that every proper consistent learning rule is probably approximately correct (PAC). Blumer, Ehrenfeucht, Haussler, and Warmuth showed, assuming the Continuum Hypothesis, that the "well-behavedness" condition of the concept class cannot be omitted: they constructed a concept class of Borel sets of VC dimension one admitting a consistent learning rule that is not PAC. We show that the Continuum Hypothesis is unnecessary. Working in Zermelo–Fraenkel set theory with the Axiom of Choice (ZFC) alone, we construct a concept class of Borel sets on $[0,1]$ of VC dimension one and a proper consistent learning rule that is not PAC. More precisely, for a suitable Borel probability measure and target concept, the rule has true risk one at every sample size on a set of samples of outer probability one. Consequently, finite VC dimension and Borel measurability of the individual concepts do not suffice to guarantee that every proper consistent learning rule is PAC. The result shows, with no need of extra set-theoretical assumptions, that the additional regularity assumption in the fundamental
theorem cannot in general be omitted.
Read the full abstract ▾
Robotics
CanonNav: Disentangling Navigation Behavior from Camera Geometry in Cross-Platform Visual Navigation
Dong-Wook Kim, Ji-Hoon Hwang, E-In Son et al. (+2 more)
While visual
navigation has advanced through imitation learning from cross-platform demonstrations, fully leveraging such data remains challenging. First, directly learning from image-trajectory pairs entangles
navigation behavior with platform-dependent camera geometry. This hinders consistent learning by forcing the policy to implicitly infer camera geometry from visual observations, an inherently ill-posed problem. Second, imitation learning from demonstrated trajectories captures the expert's chosen motion but leaves the intermediate decisions underlying that motion implicit. To address these issues, we propose CanonNav, a visual
navigation framework that disentangles
navigation behavior from camera geometry and incorporates complementary planning supervision into learning from cross-platform demonstrations. CanonNav introduces camera geometry canonicalization, which transforms visual observations and trajectories into a camera-consistent representation space. Building on this representation, we derive safety and local-progress supervision using pseudo-labels from an offline traversability
estimator. Safety supervision penalizes unsafe trajectories, while local-progress supervision guides where the
robot should advance. Experiments across diverse camera configurations and environments show that, despite using only RGB at inference, CanonNav consistently outperforms RGB-based baselines and even surpasses RGB-D-based methods in challenging scenarios.
Read the full abstract ▾
All authors Dong-Wook Kim, Ji-Hoon Hwang, E-In Son, Mintaek Oh, Seung-Woo Seo
Computation and Language
PaperBanana-Interact: Scientific Diagram Refinement with Multi-Turn Human Feedback
Xueqing Wu, Ashwin Balasubramanian, Bingxuan Li et al. (+7 more)
Recent efforts have aimed to automate scientific diagram generation from paper content (Lin et al., 2026; Zhu et al., 2026a). However, fully satisfying an author's visual and communicative preferences in a single turn is challenging: in our formative user study (N = 14), all participants requested further revisions after viewing an initial draft, and 86% of them rated the refined diagrams as more satisfactory. Despite the clear demand, the multi-turn workflow remains largely underexplored. To bridge this gap, we present MTPaperBananaBench, a
benchmark for multi-turn diagram generation containing 292 images annotated with 3,518 user requirements. To reduce expensive human studies and enable scalable benchmarking, we construct a user simulator that, at each turn, identifies unsatisfied requirements and converts k of them into natural
language feedback. Evaluating both requirement satisfaction and overall diagram quality reveals two key failure modes shared across baseline multiturn systems: (1) quality drift, where diagram quality progressively declines over turns, and (2) forgetting, where previously implemented features are lost in subsequent turns. To address these issues, we introduce PaperBanana-Interact, a multi-
agent system that refines diagrams via an internal critique-and-refine loop. PaperBanana-Interact consistently improves rather than degrades diagram quality across turns, outperforming baselines by 11.9-18.6 points in quality score and reducing forgetting by 3.7-6.2 points.
Read the full abstract ▾
All authors Xueqing Wu, Ashwin Balasubramanian, Bingxuan Li, Dawei Zhu, Kai-Wei Chang, Yale Song, Yiwen Song, Rui Meng, Tomas Pfister, Nanyun Peng
Computational Complexity
Unrestricted Boolean Multiplicative Complexity of Four-Term Binary Polynomial Multiplication: Rational Places, Hasse Jets, and the Failure of Nonlinear Feedback
Gregory Morse
Classical lower bounds show that multiplying two degree-three
polynomials over $\mathbb F_2$ requires nine scalar products in bilinear or quadratic models. They do not settle unrestricted Boolean multiplicative complexity: an XOR–AND
circuit may reuse nonlinear intermediate wires, and Boolean equality is taken modulo $x_i^2=x_i$, so a multiplication can lower algebraic degree. Let $\operatorname{Mul}_4:\mathbb F_2^8\to\mathbb F_2^7$ output the seven coefficients of the product of two four-term binary
polynomials. We prove that its unrestricted XOR–AND multiplicative complexity is exactly nine. This resolves, for a natural vector-valued quadratic function, the Boyar–Find question of whether a quadratic-
circuit lower bound can persist against unrestricted nonlinear reuse. The proof is structural rather than exhaustive. A useful purely quadratic prefix is forced onto the three rational places of $\mathbb P^1(\mathbb F_2)$. In a hypothetical eight-AND
circuit, the unique non-useful gate must carry a cubic high part. Any useful continuation then forces a rational tangent and exposes a first Hasse jet, while exterior jet separation together with Boolean idempotence prevents the same defect from exposing the second Hasse jet. The required useful suffix therefore cannot exist. A complete Lean 4 formalization verifies the Boolean-ANF semantics, the unrestricted
circuit model, and the exact
theorem; it uses no project-specific axiom or native decision procedure. The same zero-defect flag argument gives multiplicative complexity six for three-term multiplication, and the method isolates the multi-defect obstruction for five terms.
Read the full abstract ▾
Robotics
Motus2: A Self-Evolving General World Model for Dexterous Manipulation
Hongzhe Bi, Zihao Zhou, Yihang Tang et al. (+16 more)
General
embodied agents should perceive, predict, act, evaluate, and improve within a unified system. World models have shown great promise in building such
agents, yet existing models typically append an action output head to a world simulator, without coupling them into a closed decision-and-learning loop for policy improvement. We present Motus2, a self-evolving general world model for dexterous manipulation. Motus2 advances world modeling through model scaling and data scaling. For model scaling, a single model with shared weights exposes three control interfaces: a policy (world-action model), a simulator (action-conditioned world model), and an evaluator (value model). The policy proposes candidate action chunks, the simulator predicts their visual consequences, and the evaluator assesses the predicted outcomes. Their coupling forms a closed decision-and-learning loop for policy improvement. This formulation uses curated expert demonstrations for action learning, while failed and suboptimal interactions provide valuable evidence for dynamics modeling and value learning. For data scaling, Motus2 progresses from large-scale
monocular egocentric data to synchronized stereo
egocentric data, followed by
robot-domain adaptation with
robot trajectories and supplementary human-
robot alignment data. Motus2 further studies global-autoregressive and hybrid-
memory extensions of its sliding-window context, adds tactile feedback for contact-aware control, and is instantiated on a fully biomimetic platform with stereo
vision, dual arms, dual dexterous hands, and tactile sensing. Together,
egocentric data scaling and closed-loop general world model scaling provide a general path toward self-evolving dexterous manipulation.
Read the full abstract ▾
All authors Hongzhe Bi, Zihao Zhou, Yihang Tang, Jingrui Pang, Shuhe Huang, Haitian Liu, Runqing Wang, Shuai Huang, Yichen Wang, Yiming Cheng, Ruowen Zhao, Zhenghua Li, Hengkai Tan, Xiaolong Liu, Jinhui Wan, Jiabao Liu, Min Zhao, Fan Bao, Jun Zhu
Artificial Intelligence
LLM-Based Knowledge Graph Completion Combining Discrete Structural Coding with Similar Entity Information
Jiaqi Wang, Dongying Lin, Yang Yang et al. (+3 more)
Knowledge graph completion requires models to use both textual descriptions and relational structure. Existing LLM-based methods either encode KG structure as discrete
tokens or refine a restricted set of candidate entities, and these two directions have largely been studied separately. We propose CoSC for LLM-based KGC, which combines discrete structural coding with similar entity information. Specifically, an LLM generates an initial candidate entity ranking from discrete structural codes, after which information from entities with structures similar to that of the query entity refines the ranking. Experiments on FB15k-237 show that CoSC outperforms existing baselines on MRR and Hits@10 while remaining competitive on Hits@1.
Read the full abstract ▾
All authors Jiaqi Wang, Dongying Lin, Yang Yang, Yinan Liu, Bin Wang, Xiaochun Yang
Artificial Intelligence
CoLa-ICD: A Knowledge-Enhanced Framework for Long-Tail Automated Medical Coding
Yihang Cheng, Veronica Liesaputra and Andrew Trotman
Automatic medical coding assigns ICD codes to clinical notes, but it remains challenging due to long documents, imbalanced label distributions, and diverse terms. These challenges are especially severe for rare codes, which have limited training instances and are easily confused with semantically similar labels. We introduce CoLa-ICD, a knowledge-enhanced framework for long-tail prediction. CoLa-ICD enriches ICD labels with external terms, models dependencies among related codes, and learns stronger
alignment between label semantics and clinical evidence for long-tail prediction. Experiments show that CoLa-ICD improves long-tail prediction with larger gains in larger and sparser label spaces and achieves state-of-the-art performance in AUC, F1, and P@k. Our code is available at https://github.com/youwillbethebest/Cola-ICD.
Read the full abstract ▾
Computer Vision
Semantic-Spatial Discriminability Enhancement for Generalized Visual Grounding
Kaiyan Lei and Xu-Yao Zhang
Generalized Visual Grounding (GVG) task aims to localize targets in an image based on referring expressions, extends the classical visual grounding paradigm by integrating multi-target and non-target scenarios. Previous methods typically rely on global semantic matching or coarse-grained region interactions for localization, where the discriminative cues are primarily derived from sentence-level semantics or regional context. In complex multi-target scenarios, such approaches tend to confuse visually similar targets, making it difficult to establish stable instance-level decision boundaries. To address these limitations, this paper proposes a novel Semantic-Spatial Discriminability Enhancement (SSDE) framework for generalized visual grounding, which aims to enhance the discriminative ability on fine-grained semantics and spatial localization, improving both cross-modal understanding and instance-level grounding. Specifically, to enhance the semantic discriminability of query representations at the fine-grained level, we propose a Semantic Discriminability Enhancement (SeDE) module, which leverages spatially guided cross-
attention to disentangle fine-grained target-relevant visual attributes and integrates them with the textual subject semantics. Furthermore, to strengthen the spatial discriminability of the referred targets, we introduce a Spatial Discriminability Enhancement (SpDE) module, which models an instance center density map to characterize the spatial distribution of targets, and explicitly constructs instance separation structures in the spatial domain by employing them as an auxiliary supervision signal. Extensive experiments show that SSDE achieves superior performance on ten
datasets across both classic and generalized visual grounding tasks.
Read the full abstract ▾
Formal Languages and Automata · Cross-listed from q-bio.NC
"More Is Different” in Neural Circuits: Algebraic Emergence of Effective Theories in Canonical Recurrent Motifs of Biological Neuronal Networks
Nima Dehghani
Canonical neural
circuit motifs are usually described functionally: divisive normalization rescales population activity by a pooled signal, and winner-take-all competition selects one pattern through recurrent excitation and shared inhibition. We represent them, and their compositions, algebraically as finite transformation systems and analyze the transition monoids generated by their input-conditioned updates, distinguishing structure already present in a generator from structure that appears only through composition, and, on a joint state space, structure inherited from one factor from structure that lives on a joint configuration. Individually aperiodic updates can generate non-aperiodic monoids. In the WTA, every frozen-drive generator collapses to fixed points, yet short input sequences create local cycles of winner-dependent inhibitory gating: globally dissipative dynamics with a reversible action. The strongest result arises in WTA-to-DN composition. The composed monoid then contains a genuinely composite local cycle in which normalization state and the winner's gating state change together, although every primitive generator is aperiodic. Holonomy analysis certifies this as a group component of the Krohn-Rhodes cascade rather than an incidental cycle, and finds most group-carrying image sets on joint configurations, whereas the uncoupled product has none. An exhaustive interface sweep shows that the composite cycle is a property of the coupling rather than of a chosen map. If motifs are building blocks of neural computation, composing them is a form of programming: one chooses primitives and interfaces so that the generated
algebra has the intended repertoire. The transition monoid is that repertoire - what a primitive presents to any later construction. Recurrent
circuits are compositional transformation systems; their
algebra constrains what they can be programmed to compute.
Read the full abstract ▾
Artificial Intelligence
Rethinking the Test-Time Prompt Tuning Objective from the Perspective of Calibration
Jungwon Choi, Hyeonseo Jang, Kibok Lee et al. (+1 more)
Test-time
prompt tuning (TPT) has emerged as a powerful paradigm, refining
prompts for each test sample via
entropy minimization (EM) over multiple augmented views. However, we identify a limitation in the standard EM-based adaptation: it inherently drives the model toward overconfident predictions disregarding sample-specific uncertainty, leading to significant calibration degradation. To address these limitations, we propose a new objective that replaces the conventional EM loss by aligning the original-view prediction with a target distribution derived from augmented views via cross-
entropy, while adversarially incorporating the
entropy of the target distribution to capture sample-specific uncertainty. Furthermore, to better construct this target distribution, we apply confidence-aware temperature scaling to each augmented-view prediction according to its confidence, sharpening confident predictions while softening uncertain ones. This formulation allows the model to increase confidence only when the target distribution is reliable, while preserving uncertainty when it reflects ambiguous or conflicting augmented-view predictions. Extensive experiments across diverse
benchmarks demonstrate that our approach not only achieves state-of-the-art accuracy but also significantly improves model calibration.
Read the full abstract ▾
All authors Jungwon Choi, Hyeonseo Jang, Kibok Lee, Eunwoo Kim
Computation and Language
Quantifying and Mitigating Korean Jamo-Level Typographical Vulnerabilities in Large Language Models
Seojin Lee and Hwanhee Lee
Korean introduces an additional typographical perturbation level not captured by ordinary character-level edit models: because syllable blocks are internally composed of sub-character units called jamo, keyboard-level errors can occur within a syllable, either producing a valid but semantically altered character or exposing raw jamo on the surface. Both outcomes disrupt sub-word tokenization and are not reliably corrected by existing grammatical error correction pipelines, leaving LLMs directly exposed to corrupted inputs. To quantify this vulnerability, we apply five jamo-level perturbation types to the KMMLU
benchmark and evaluate four
language models, finding that accuracy declines monotonically with perturbation intensity and that parameter scaling does not confer robustness against intra-syllabic noise. We further show that typo-corrupted inputs induce a distinct shift in internal representations that is not reducible to ordinary answer incorrectness, and that a simple linear probe trained on these representations detects unseen perturbation types with high AUROC. Motivated by this signal, we propose Typo-Aware Chain-of-Thought (TACoT), which routes inputs to chain-of-thought inference only when the probe detects a likely typo, recovering a substantial portion of the CoT accuracy gain at a fraction of the inference cost.
Read the full abstract ▾
Databases
ELASTIC: Trajectory-Based Synchronization of Event and Tracking Data in Soccer
Hyunsung Kim, Hoyoung Choi, Kunhee Lee et al. (+4 more)
Combining event and tracking data is fundamental to modern soccer analytics, yet the two sources are rarely well aligned: event timestamps recorded by human annotators often miss the true moment of the action, distorting the spatiotemporal context that downstream models rely on. Existing synchronization methods depend on noisy human-annotated event locations and fail to detect ball receptions, obscuring when each player gains ball possession. To address these limitations, we propose ELASTIC (Event-Location-AgnoSTIC synchronizer), a framework that infers the start and end timestamps of events solely from player and ball trajectories, without relying on annotated event locations. To recover ball receptions, ELASTIC enriches the event sequence by inserting virtual termination events between consecutive events, so that the end of each event is detected jointly with its start. It then extracts a sparse set of candidate frames where ball touches are physically plausible, and aligns the termination-inserted event sequence with the candidate-frame sequence using an extended Needleman-Wunsch
algorithm. For reproducible evaluation, we construct a publicly available
benchmark by annotating ground-truth timestamps on the Sportec Open DFL
Dataset, on which ELASTIC substantially outperforms existing methods. Through downstream task evaluation, we further show that improved synchronization translates into measurable gains in soccer analytics. The source code and
benchmark are available at https://github.com/hyunsungkim-ds/elastic.git.
Read the full abstract ▾
All authors Hyunsung Kim, Hoyoung Choi, Kunhee Lee, Sangwoo Seo, Tom Boomstra, Jinsung Yoon, Chanyoung Park
Artificial Intelligence
LaMoC: Loss-Aware Modular Compression for LLMs
Mohanad Odema and Jacob Song
Modular compression has enabled considerable parameter reduction in LLMs while preserving strong
language understanding and downstream task accuracy. However, existing joint modular compression methods primarily rely on activation statistics, leaving loss-sensitivity information and its module-level characterization underexplored. We investigate addressing this gap with LaMoC, a loss-aware modular compression methodology that blends activation and Empirical Fisher statistics through gradient-error
alignment. LaMoC improves joint compression by selecting compression statistics that better align local module reconstruction error with the downstream loss. Our contributions are three-fold: (1) We characterize the Empirical Fisher as a module-level loss-aware proxy that can be blended with the activation statistics required for compression. (2) We reformulate joint modular compression as a two-tiered optimization problem that minimizes module reconstruction error while tuning the activation and gradient information blending rate. (3) We implement an empirically driven methodology with statistical validation to solve the resulting compression problem. We evaluate LaMoC across four model families spanning eight models. On the 4-8B models, LaMoC achieves an average 2.5% reduction in perplexity and a 1% relative improvement in task accuracy over state-of-the-art modular compression methods.
Read the full abstract ▾
Computational Engineering
Redefining Stablecoins from Nominal to Real Value: A Maximum Likelihood Approach
Tomonori Kanno, Kensuke Ito, Yushi Yoshimura et al. (+1 more)
Stablecoins, typically pegged to fiat currencies, cannot achieve true stability because they inherit fluctuations in the underlying unit of account. To overcome this limitation, we introduce a stablecoin pegged to the Maximum Likelihood Value (MLV), a newly defined unit of account derived as the most probable configuration of latent real-value movements that explains observed nominal-value (price) changes. Grounded in inferential statistics and modern portfolio theory, MLV represents the most stable unit of account, as it enforces a zero real return on the minimum-variance portfolio. Empirical results confirm the operational viability of an MLV-pegged stablecoin: MLV can be computed in real time from 500 asset price series and improves annualized returns and Sharpe ratios while substantially reducing turnover in portfolio optimization.
Read the full abstract ▾
All authors Tomonori Kanno, Kensuke Ito, Yushi Yoshimura, Kyohei Shibano
Computation and Language
The Differential Reasoning Router: Operationalizing Cost-Aware LLM Annotation in E-commerce
Cheng Lyu, Jingyue Zhang, Vinny DeGenova et al. (+2 more)
Large
Language Models (LLMs) are increasingly used to annotate structured product data in e-commerce, but early deployment often begins as a cold-start problem: only limited pre-launch labels are available, the value of expensive
reasoning is unknown, and human review is needed before the system can be trusted at scale. This challenge is especially common in rule-based annotation workflows, where each item must satisfy multiple business rules and both model errors and ambiguous rule boundaries affect final decisions. We introduce the Differential
Reasoning Router (DRR), a cost-aware framework for cold-start LLM annotation that jointly optimizes model selection and human escalation. Rather than treating a
reasoning model as a default fallback, DRR estimates separate success probabilities for a direct model and a
reasoning model at both the sample and business-rule levels, enabling adaptive routing: easy cases are handled directly,
reasoning is reserved for cases where it is expected to improve the decision, and likely double-failure or rule-disagreement cases are escalated to human annotators. The resulting labels provide targeted ground truth for
prompt engineering, supervised fine-tuning, calibration, and rule refinement, enabling a gradual shift from human-heavy cold-start annotation toward high-confidence automated routing. In a production e-commerce workflow, DRR reaches accuracy parity with the strongest confidence-based router while achieving more than 60%
reasoning-
token cost savings.
Read the full abstract ▾
All authors Cheng Lyu, Jingyue Zhang, Vinny DeGenova, Mengwei Li, Yuanli Pei
Machine Learning · Cross-listed from stat.ML
Fairness in multi-class multi-group classification problems via contextial coherent risk measures
Darinka Dentcheva and Xiangyu Tian
We propose a new design of fair classifiers for multi-class classification problems in the presence of vector-valued sensitive attributes. In that scenario each sensitive attribute has multiple values and forms several groups relevant to the
fairness consideration. Naturally those groups are overlapping and one should also analyze the interaction of factors. Additionally, the decision makers aided by the classification should not violate individual rights at the expense of satisfying
fairness metrics at the group level. We propose an approach using the theory and methods of coherent measures of risk aiming at resolving the
fairness challenges. Further, we propose a specialized numerical method for solving the resulting optimization problem. The method scales well with the increase of the number of observations. Additionally, we note that the obtained classifier is robust with respect to corrupted data or to situation when data is scarce. We demonstrate the advantages of the proposed framework in comparison to the support-vector machine framework and other methods handling
fairness.
Read the full abstract ▾
Robotics
Optimized Modular Design and Development of a Tilt-Rotor Bicopter Drone
Saideep Verma, Nimisha Tatapudi, Akshay Arjun et al. (+3 more)
Hybrid systems like tilt-rotor bicopter drones combine the beneficial characteristics of both fixed-wing and rotary-wing technology, enabling long endurance and VTOL capability. However, such drones also require an optimum design to ensure both static and dynamic stability. The modular design of a traditional bicopter is developed in this paper based on extensive analysis and in-depth structural and aerodynamic simulations. The structural analysis has been performed to ensure that the aircraft's structure withstands the stresses encountered during different flight modes. Controlling the relative positions of the Center of
Gravity (CG) and Neutral Point (NP) is an essential aspect of the design, ensuring stability during hover and positive stability during forward flight. The thrust and power analyses have been conducted to assess the flight performance and endurance. After analysis, the drone has been developed, and flight tests with a basic flight controller were conducted to validate the performance metrics obtained in the simulation.
Read the full abstract ▾
All authors Saideep Verma, Nimisha Tatapudi, Akshay Arjun, Danwada Sharanya Sukka, Abhishek Sarkar, Joyjit Mukherjee
Robotics
Contrast-Free Autonomous Navigation of Untethered Endovascular Microrobots Using Single-Plane Fluoroscopy
Husnu Halid Alabay, Tuan-Anh Le, Ping Wang et al. (+1 more)
Reliable three-dimensional (3D)
navigation of magnetically actuated untethered microrobots remains a major barrier to clinical translation. X-ray fluoroscopy is the standard real-time imaging modality for endovascular procedures, but single-plane fluoroscopy provides only a two-dimensional (2D) projection, eliminating depth information and complicating autonomous
navigation. Recovering this information through biplane imaging or repeated contrast-enhanced angiography increases procedural complexity, radiation exposure, or contrast burden. Here, we introduce VISTA (Virtual Integration for Spatial Tracking and Autonomy), a digital twin framework enabling contrast-free autonomous
navigation under single-plane fluoroscopy. VISTA reconstructs vascular anatomy as a 3D digital twin, discretizes the vessel centerline into
navigation milestones, and assigns the detected 2D
robot position to the nearest projected milestone. Consecutive milestones define the local vessel orientation used to generate
magnetic actuation commands, converting single-plane fluoroscopic observations into
topology-constrained
navigation states without requiring contrast injection during
navigation. VISTA is demonstrated across anatomically distinct vascular phantoms under continuous flow and within the inferior vena cava of a live rat in vivo. Compared with conventional fluoroscopic human-in-the-loop control, VISTA reduced
navigation time by up to 62%, corrective actuation commands by up to 98%, and radiation exposure by up to 57%. These results establish VISTA as a digital twin-guided framework for contrast-free autonomous
navigation of untethered endovascular microrobots using widely available single-plane fluoroscopy.
Read the full abstract ▾
All authors Husnu Halid Alabay, Tuan-Anh Le, Ping Wang, Hakan Ceylan
Computer Vision
Amortized Anchor Refinement for Deployable Continuous-Time 4D Gaussian Reconstruction
Jingong Chen, Qingwen Zhang, Sanghyeon Jun et al. (+3 more)
Continuous-time 4D reconstruction remains impractical on standalone XR headsets. Per-scene optimization demands deployment-infeasible compute, and lower budgets cause collapse rather than degrade gradually. Feed-forward prediction is fast, but struggle to recover scene-specific detail. We present Amortized Anchor Refinement, which uses a frozen backbone to predict an initial Gaussian representation and a short optimization to specialize it under a fixed compute budget, with a capacity floor preserving representational density. A training-free stage then applies a persistent-homology constraint to prune unstable Gaussians while preserving topologically persistent structures, and streams the resulting trajectories directly as scene flow. On the Stage-Capture
benchmark, Amortized Anchor Refinement achieves 24.31$\pm$2.22dB, while our deployment experiments demonstrate reconstruction within the target budget on a single consumer GPU and playback on a standalone XR headset.
Read the full abstract ▾
All authors Jingong Chen, Qingwen Zhang, Sanghyeon Jun, Chulwoo Pack, Kyle Gao, Kwanghee Won
Computation and Language
Label Semantic Expansion via Label Guided Neural Topic Modeling
Haojia Zheng, Yuyin Lu, Juntian Huang et al. (+4 more)
Topic models are widely used for content analysis, where users often analyze corpora around predefined labels rather than unordered latent topics. Existing label-aware topic models mainly follow a labels-for-topics perspective, using labels to guide topic learning, while the learned topics are not directly usable for label-centered analysis. We explore the reverse topics-for-labels perspective and instantiate it as Label Semantic Expansion (LSE), which enriches sparse label representations with corpus-grounded descriptive topic words. To exploit topics in LSE effectively, we propose a Label-Guided Neural Topic Model (LGNTM), which learns dedicated label-aligned topics, grounds them in lexical and document semantic spaces, and preserves consistency between topic structures and label structures. Experiments on label-topic
alignment, label expansion, topic quality, and downstream classification demonstrate strong overall performance across complementary evaluation dimensions.
Read the full abstract ▾
All authors Haojia Zheng, Yuyin Lu, Juntian Huang, Fan Ou, Yanghui Rao, Haoran Xie, Fu Lee Wang
Systems and Control
Control of Decommissioned Satellites and Space Debris Using CubeSats with Ion Electrospray Engines
Felix Biertümpfel, Peter Seiler, Paulo Lozano et al. (+1 more)
The emergence of the New Space era has led to a rapidly increasing number of satellites in Low Earth Orbit (LEO). Consequently, more stringent deorbiting requirements have recently been imposed to avoid the Kessler syndrome in LEO. This has resulted in the proposal of new concepts for space debris removal, including attaching CubeSats to space debris as a promising mitigation strategy. The initial phase of this strategy involves stabilizing and controlling the debris' attitude. This paper proposes an attitude control design for decommissioned satellites using attached CubeSats with staged ion electrospray engines (iESE). The compact design of iESE combined with staging provides increased reliability and mission durations. The large uncertainty in the dynamics of the combined system, i.e. satellite and attached CubeSats, poses a significant challenge to the control design. The approach taken here uses the robust control framework, specifically $\mu$-synthesis, to tackle this challenge. The feasibility of the approach is demonstrated on a decommissioned satellite with multiple flexile appendages.
Read the full abstract ▾
All authors Felix Biertümpfel, Peter Seiler, Paulo Lozano, Harald Pfifer
Artificial Intelligence
SPARK: Skeleton-Guided Reasoning Synthesis from Large-Scale Scientific Literature
Yu Li, Wei Li, Xin Gao et al. (+4 more)
Scientific
reasoning remains challenging for open-source models, largely due to the lack of high-quality scientific
reasoning data. Existing
datasets are often dominated by factual recall or formulaic problem solving, with limited emphasis on mechanism understanding, evidence-grounded
reasoning, and hypothesis evaluation. To address this, we introduce SPARK (Scientific Paper Abstracted
Reasoning sKeleton), a paper-oriented synthesis framework built on Sci-Base, a large-scale corpus of research papers spanning 10 scientific disciplines. Instead of directly converting papers into question-answer pairs, SPARK treats the claim-evidence-derivation structure of a paper as the fundamental unit of
reasoning synthesis. Specifically, SPARK (1) distills each paper into a compact
reasoning skeleton capturing its central claims and supporting evidence, enabling self-contained question generation, and (2) synthesizes
reasoning tasks from four scientific perspectives: mechanistic
reasoning, hypothesis falsification, quantitative derivation, and boundary calibration. A final consistency verification stage further removes unsupported or contradictory outputs. Using this framework, we construct Spark-234K, a scientific
reasoning dataset with substantially higher difficulty and diversity than existing resources. Experiments show that Spark-234K consistently outperforms existing scientific
reasoning datasets while achieving stronger performance with significantly fewer training samples.
Read the full abstract ▾
All authors Yu Li, Wei Li, Xin Gao, Mengyuan Sun, Xiaoyang Wang, Qizhi Pei, Lijun Wu
Computer Vision
Towards a Joint Khmer Text Recognition and Word Segmentation
Marry Kong, Rina Buoy, Sovisal Chenda et al. (+3 more)
Text recognition, or extracting electronic text from document images, has been indispensable for knowledge
retrieval tasks, such as
retrieval-augmented generation (RAG). For Khmer, extracted text is subject to an extra word
segmentation step, as Khmer does not use any visible word delimiters to denote word boundaries. Thus, a recognition-then-
segmentation pipeline for Khmer requires two separate sequential models; this is not only error-prone but also adds significant latency for large-scale document processing. This paper proposes a novel joint Khmer text recognition and word
segmentation framework in a unified model. The proposed model, using a connectionist-temporal-classification (CTC) decoder for fast, parallel decoding, can be instructed to recognize Khmer text with ($b=1$) and without ($b=0$) word
segmentation. Experimental results on different
benchmark datasets of different document modalities (document, scene, and handwritten images) show that the proposed model can not only recognize characters in document images but also locate word boundaries, removing the need for an extra word
segmentation step in a conventional sequential pipeline.
Read the full abstract ▾
All authors Marry Kong, Rina Buoy, Sovisal Chenda, Nguonly Taing, Masakazu Iwamura, Koichi Kise