Artificial Intelligence
Training LLMs for Multi-Agent Collaboration with Verifiable Information Rewards
Keru Chen, Sen Lin, Yingbin Liang et al. (+2 more)
Effective collaboration among large
language model (LLM)
agents in multi-
agent systems requires sharing relevant evidence, requesting missing information, and incorporating teammates' contributions into decisions. Training these behaviors is challenging because final task outcomes provide limited feedback on the value of individual messages. We introduce CoSIR (Collaboration via Surprisal-based Information Rewards), a post-training approach that uses a simple, verifiable environment to teach collaboration behaviors that generalize to more complex, unseen tasks. We propose a
Bayesian evaluator that scores shared information by its reduction in team surprisal, defined as the sum of the negative log probabilities assigned to the correct option given each
agent's available evidence. Requests are scored by the potential reduction in the same quantity if the requested information were truthfully provided. Separate rewards encourage
agents to use the received information in their private choices and calculations. We post-train Qwen3-4B, Qwen3-8B, and Gemma-3-4B-it through self-play and evaluate transfer to seven unseen natural-
language collaboration tasks. CoSIR achieves the highest aggregate score, while its performance in general tasks remains competitive. Behavioral analyses show increased disclosure of private evidence and fewer judged failures to use teammates' input.
Read the full abstract ▾
All authors Keru Chen, Sen Lin, Yingbin Liang, Nathaniel D. Bastian, Shaofeng Zou
Numerical Analysis
Provable Method of Moments for Cryo-EM
Amnon Balanov, Tamir Bendory, Dan Edidin et al. (+3 more)
We study the recovery of a three-dimensional volume in single-particle cryo-electron microscopy (cryo-EM) from noisy tomographic projections acquired at unknown viewing rotations, drawn independently and uniformly from $\mathrm{SO}(3)$. We show that, in a spherical-shell representation satisfying an explicit shell-separation condition, the first three moments generically determine the underlying real volume up to its intrinsic global $\mathrm{O}(3)$ ambiguity. The proof is constructive and yields a frequency-marching procedure that successively recovers the volume. Consequently, consistent recovery in the high-noise regime has sample complexity $n=\omega(\sigma^6)$, where $n$ is the number of observations and $\sigma^2$ is the noise variance. Numerical experiments validate the constructive recovery scheme under finite sampling and additive noise, and indicate that direct third-moment optimization can succeed beyond the structural assumptions required by the constructive proof.
Read the full abstract ▾
All authors Amnon Balanov, Tamir Bendory, Dan Edidin, Josh Katz, Shay Kreymer, Nir Sharon
Combinatorics
The logarithmic exponent of the Ramsey number $r(4,t)$
Qizhong Lin and Qiuyang Zhang
We prove that $r(4,t) \ge \frac{t^3}{(\log t)^{2+o(1)}}$. This improves the lower bound of Mattheus and Verstraëte (Ann. of Math., 2024) by a factor of $(\log t)^{2-o(1)}$.
Symplectic Geometry
Metaplectic-c prequantizations and the Einstein–Brillouin–Keller quantizability condition
Hyunmoon Kim
We show that every symplectic
manifold admits a symplectic form whose metaplectic-c prequantizations give quantizability conditions on all its Lagrangian submanifolds. Metaplectic-c prequantizations descend to prequantizations over the Lagrangian Grassmannian bundle that are fibrewise flat with holonomies primitive fourth roots of unity on the generators of the fundamental groups. This fibrewise holonomy is consistently interpreted as that of both Leray two-orientations and normalized Blattner–Kostant–Sternberg half-forms. The prequantizations pull back to flat circle bundles on each Lagrangian submanifold whose holonomy character defines a quantizability condition. The first and second Stiefel–Whitney classes are identified as splitting obstructions to the Lagrangianwise holonomies. These holonomies split into a product of holonomies on any orientable Lagrangian submanifold, and split consistently for all Lagrangian submanifolds of a
manifold with metaplectic structure. When the first Chern class vanishes, the holonomy splits into the Bohr–Sommerfeld factor and a quarter phase Maslov correction, recovering the classical Einstein–Brillouin–Keller condition. We develop the underlying obstruction theory for circle extensions associated to general Lie group double coverings with first hypercohomology groupoids.
Read the full abstract ▾
Earth and Planetary Astrophysics
Size, atmospheric, and ring constraints on the large Kuiper belt object (55565) Aya from a stellar occultation
Ko Arimatsu, Hidekazu Hanayama, Satoshi Takita et al. (+3 more)
We aim to constrain the projected size,
geometric albedo, possible atmosphere, and possible ring material of the mid-sized trans-Neptunian object (55565) Aya from the stellar occultation of 2024 May 10 (UT). We observed the event from three Japanese sites with high-cadence optical imaging. We combined one positive occultation chord with an informative prior on the shadow centre derived from orbital and astrometric uncertainties, and inferred the body radius with
Bayesian Markov chain Monte Carlo modelling. We constrained a possible atmosphere by fitting refraction models for pure CH$_4$, N$_2$, and CO atmospheres to multi-band light curves, and searched for ring-like extinction signatures in unbinned and binned high-time-resolution photometry. We derive an equivalent radius, i.e., the radius of the circle fitted to the projected shadow, of $r=401^{+28}_{-40}$ km for Aya, corresponding to a
geometric albedo of $p_V=0.101^{+0.023}_{-0.013}$. This estimate is insensitive to the adopted ephemeris solution and is consistent with an ephemeris-independent analysis that makes no assumption about the position of the shadow centre. We obtain 3$\sigma$ upper limits on the atmospheric surface pressure of $p_0=60$, 90, and 80 nbar for pure CH$_4$, N$_2$, and CO compositions, respectively. No significant ring-like extinction is detected over projected distances of about 430-8060 km sampled by the three stations, and our most sensitive data exclude dense narrow features with apparent opacity $p'\gtrsim0.1$ and radial widths of a few kilometres or larger over $\sim3r$-$14r$. This is the first stellar occultation by Aya ever observed. In addition to constraining Aya's size and albedo and finding no ring-like extinction, we place a stringent atmospheric upper limit that supports diversity in atmospheric states among similarly sized trans-Neptunian objects.
Read the full abstract ▾
All authors Ko Arimatsu, Hidekazu Hanayama, Satoshi Takita, Fumi Yoshida, Tsutomu Hayamizu, Jun-ichi Watanabe
Representation Theory
Multiplicity One for Five Strongly Tempered Spherical Varieties
Yuxuan Ge
Let $F$ be a non-Archimedean local field of characteristic zero. We prove multiplicity one for every irreducible smooth admissible representation in the split forms of Wan-Zhang Models 6-10. Combining the Gelfand-Kazhdan method with nilpotent-wavefront restrictions on Bernstein-finite equivariant distributions, we establish the required
symmetries through rational orbit analysis, parabolic descent, stabilizer characters, and conormal arguments. The type-$D_6$ analysis treats both the vector-similitude group of Model 9 and the semispin form arising in the exceptional $E_7$ model. For Model 10, the distribution
theorem gives a product bound for periods; the individual multiplicity bound follows from Langlands classification and a vanishing
theorem for standard modules induced from nongeneric tempered data on the maximal $E_6$ Levi.
Read the full abstract ▾
Databases
Path Length Existence Problems in Bounded-Treewidth Probabilistic Graphs
Antoine Amarilli, Mikaël Monet and Harshul Sagar
We consider the problem of computing the probability of existence of a source-to-target path or walk satisfying a given length constraint in an input probabilistic graph. The length constraints we consider can require the length to be at least $\ell$, at most $\ell$, or exactly $\ell$, for $\ell$ a target length given as input. The probabilistic graphs that we consider are directed graphs where each edge is annotated with an independent probability of existence. As this task is generally intractable (#P-hard), we aim for
algorithms applying to input graphs that are assumed to be of bounded treewidth. We classify the complexity landscape of these problems: - We give an XP
algorithm parameterised by treewidth to compute the probability that an input probabilistic graph contains a source-to-target path/walk of length $\geq \ell$, and do the same for a path/walk of length $\leq \ell$. - We show that the four problems mentioned in the last item are W[1]-hard parameterised by pathwidth (hence not FPT under the standard assumption that FPT$\neq$W[1]). This is in contrast to the case of non-probabilistic graphs, where the corresponding problems were known to be FPT parameterised by treewidth; and to the connectivity problem (and more generally MSO evaluation) on probabilistic graphs, where the same was known. - Surprisingly, we show that it is #P-hard to compute the probability that an input graph contains a source-to-target path/walk of length exactly $\ell$, even on graphs of pathwidth $2$. By contrast, all our problems are tractable on graphs of treewidth $1$. - Last, we investigate the complexity of our problems when the edges can carry weights. This causes all our problems to become #P-hard already on pathwidth-2 graphs, and causes specifically the “exactly $\ell$” walk problem to be #P-hard even on pathwidth-1 graphs.
Read the full abstract ▾
Programming Languages
Compositional Information Geometry of Probabilistic Programs
Jacek Karwowski, Frank Nielsen and Sam Staton
We study the compositional, intensional information geometry of probabilistic programs. For a first-order probabilistic programming
language with finite random variables, we construct a contravariant lax monoidal functor that, for a program of type $X \to Y$, pulls back a (degenerate) Riemannian metric on $Y$ to $X$. It specialises to the Fisher metric in the case of standard statistical models and to ordinary pullback in the case of deterministic programs. We further justify this construction in two ways. First, we show that this functor arises as a pullback of geometry from a universal space of programs over a type $Y$ with a fixed trace space. Second, we show that it is a special case of a more general pullback construction over weighted measures. Our compositional geometry depends on the presentation of the program, and we study how it relates to the (non-compositional) geometry of the realised measures by establishing a three-level inequality relating our score-velocity metric for the program, the program's corresponding output-conditioned metric, and the standard Wasserstein-Fisher-Rao metric of the realised family. Finally, we show how the theory can be extended to accommodate continuous random variables using diffeological quasi-Borel spaces. We also discuss several concrete directions for future work.
Read the full abstract ▾
Machine Learning
Deep Mesoscopic Dynamics
Yue Zhao, Shiqi Wu, Yihang Ouyang et al. (+6 more)
Microscopic simulations resolve spatial patterns but are feasible only over small domains and short timescales, whereas bulk descriptions erase those patterns through spatial averaging. Mesoscopic fields preserve this spatial organisation, but their effective equations are known only for selected microscopic systems and scaling limits. We introduce Deep Mesoscopic Dynamics (DeepMesoD), which uses neural networks to learn an effective mesoscopic field equation from microscopic trajectories at a prescribed coarse-graining scale. For the nonconserved dissipative dynamics studied here, DeepMesoD learns the free-energy functional with a spatial interaction kernel and a state-dependent dissipation operator constrained by the fluctuation-dissipation relation. We first validate the construction in the Ising-Kac model, where the learned mesoscopic equation recovers the known hydrodynamic limit and rollouts remain stable far beyond the training horizon. We then treat two systems without closed-form mesoscopic descriptions. For a two-dimensional Heisenberg ferromagnet governed by atomistic
spin dynamics, DeepMesoD learns a mesoscopic model whose predictions match atomistic references up to 1 ns, and a rollout with an unseen initial condition remains bounded up to 25 ns. For layered CrBr$_3$, the learned kernel captures the suppression and sign reversal of interlayer coupling at a stacking fault whose location and strength are not supplied as inputs. Controlled changes to the learned interaction show how fault number and placement shape multistep
magnetic responses, complementing our magneto-photoluminescence measurements. Across the two atomistic systems, DeepMesoD achieves a three-to-four-order-of-magnitude reduction in wall-clock time compared with atomistic
spin dynamics.
Read the full abstract ▾
All authors Yue Zhao, Shiqi Wu, Yihang Ouyang, Sergey Grebenchuk, Magdalena Grzeszczyk, Xiaoxu Zhao, Kostya S. Novoselov, Maciej Koperski, Qianxiao Li
Combinatorics
The chromatic number of $G$-stable Kneser graphs
Carmen Arana, Penny Haxell, Tomáš Kaiser et al. (+1 more)
Extending the notions of Kneser graphs and Schrijver graphs, we define the $G$-stable Kneser graph $KG(G,k)$ (where $k\geq 2$ and $G$ is a graph on $n\geq 2k$ vertices) as the graph whose vertices are the $k$-element independent sets in $G$ and whose edges join pairs of disjoint independent sets. The well-known result of Lovász on the chromatic number of Kneser graphs implies that $\chi(KG(G,k)) \leq n-2k+2$ for any $G$. We call $G$ $k$-tight if equality holds, and we study the following question: Which graphs are $k$-tight? A result of Daneshpajouh et al.\ implies a characterization of $2$-tight graphs. We prove a necessary condition for a graph to be $k$-tight for $k\geq 3$ which extends this characterization, and we
conjecture it to be sufficient. Another result gives a lower bound to the chromatic number of the $G$-stable Kneser graphs when $G$ is a disjoint union of cycles. We use this result to give a short proof of the fact that $4$-regular graphs on $3k$ vertices consisting of two vertex-disjoint cycles and $k$ cycles of length $3$ admit an independent set of size $k$.
Read the full abstract ▾
All authors Carmen Arana, Penny Haxell, Tomáš Kaiser, Matěj Stehlík
Mathematical Physics
Dynamics of low energy wave packets in the one-dimensional quantum Heisenberg ferromagnet
Mateusz Kwaśnicki, Marcin Napiórkowski, Sören Petrat et al. (+1 more)
Spin wave theory plays a central role in the analysis of low-energy properties of the
quantum ferromagnetic Heisenberg model. In this work, we present a first step towards a rigorous justification of this approximation in the context of
quantum dynamics in one dimension. According to
spin wave theory, the
evolution of a
spin wave state should be well approximated by its free
evolution with the appropriate energy of free quasiparticles called
magnons. Our main result establishes this idea rigorously for a class of
spin wave packets. We focus on
spin-$\frac{1}{2}$ systems, and consider wave packets composed of two
magnons. We argue our result is optimal with respect to the timescales.
Read the full abstract ▾
All authors Mateusz Kwaśnicki, Marcin Napiórkowski, Sören Petrat, Michał Pietruszka
Computer Vision
Diagnose-to-Generate: Eliminating Structural Distortions in Flow Models via Patch-Level Grounded Alignment
Yifan Li, Lishuai Gao, Tong Zhang et al. (+8 more)
Flow-based image generators achieve strong semantic synthesis but still produce localized structural distortions, such as malformed hands and limbs. Existing post-training methods typically optimize global scalar rewards, which provide weak spatial credit assignment and may alter well-formed regions while correcting local defects. We present Diagnose-to-Generate (D2G), a grounded
alignment framework that separates visual diagnosis from generative optimization. We construct Achilles-600K, a region-annotated
dataset for structural anomalies, and train ARGUS through supervised cold-start and Region-Aware Group Relative Policy Optimization to predict anomaly categories and bounding boxes. These diagnoses are converted into confidence-weighted spatial rewards for our Patch-level Flow-GRPO. We further identify that global
stochastic exploration can shift semantic regions across group rollouts, making patch-wise relative advantages spatially inconsistent. Our Trajectory-Preserving
Stochastic Sampling (TPSS) anchors early denoising with deterministic ODE trajectories and introduces stochasticity only at later steps, thereby limiting intra-group spatial drift while retaining local exploration. Experiments on FLUX.1-dev and SD3.5 show improved structural fidelity with preserved FID and CLIP scores. Consistent gains under HumanRefiner, zero-shot GPT-4.1, and blinded human evaluation further demonstrate that the improvements extend beyond the training reward model.
Read the full abstract ▾
All authors Yifan Li, Lishuai Gao, Tong Zhang, Qilong Huang, Chen Zhao, Zhuoliang Kang, Jie Hu, Yujie Zhong, Fenghe Tang, S Kevin Zhou, Xiaoming Wei
Robotics
Agentic Robot Creation through Generation, Verification, and Revision
Xuebing Li, Yibo He, Sitong Liu et al. (+2 more)
Can an AI
agent turn high-level user intent into a functioning
robot and teach it how to behave? We study this question from two complementary aspects: embodiment and behavior creation. For embodiment creation, we investigate how well current
multimodal models capture mechanical requirements, and introduce computational metrics for evaluating generated designs before fabrication. For behavior creation, we present Roboforge, an agentic framework that turns task specifications into executable policies through motion generation, policy learning, and execution-grounded revision. Given a
robot embodiment and task, the
agent generates a reference motion, configures the learning objective, and trains a policy. Crucially, motion remains in the loop: rollout feedback can revise the reference motion, reward, or both. We evaluate a generated embodiment across multiple MLLMs and behavior creation across four diverse
robots against reward-refinement and one-shot motion baselines. Our computational metrics expose recurring mechanical failures in generated designs, while behavior experiments show that execution-grounded refinement improves task-level behavior and execution quality. Finally, we fabricate an
agent-generated Gorilla and directly deploy learned behaviors on the physical
robot.
Read the full abstract ▾
All authors Xuebing Li, Yibo He, Sitong Liu, Ran Ji, Jiayuan Gu
Computation and Language
Cross-Lingual Transfer of Evil Vectors Is Directionally Asymmetric
Juann Kim and Minku Kang
Persona vectors are linear directions in the residual-stream activations of LLMs that encode behavioral traits and enable fine-grained control over model behavior. Among them, the evil persona vector is particularly relevant to understanding harmful behavior, jailbreak vulnerabilities, and the unlearning of harmful knowledge. Yet whether this vector reflects a genuinely cross-lingual concept, rather than an artifact of its extraction
language, remains untested. We conduct a systematic evil-vector transfer study across two high-resource
languages (English, Spanish) and two low-resource
languages (Swahili, Telugu) on four open-source instruction-tuned LLMs, and find that transfer is directionally asymmetric. Evil vectors are harder to obtain reliably from low-resource
languages, and the resulting vectors are largely ineffective at steering, even within their own
language. In contrast, vectors extracted from high-resource
languages transfer effectively to low-resource inputs while largely preserving general knowledge, as measured by MMLU.
Geometric analysis further supports this pattern: evil vectors from the two high-resource
languages are closely aligned, whereas low-resource vectors align only weakly with this shared direction. These results suggest that a shared cross-lingual evil direction exists and, once reliably extracted from a high-resource
language, can be reused to steer behavior in low-resource
languages without additional per-
language data collection.
Read the full abstract ▾
Superconductivity
Imaging atomic-scale surface structure and superconducting gap of the misfit layered compound (SnS)1.17NbS2
Dongming Zhao, Shamarita Deb, Lichen Ji et al. (+8 more)
Superlattices of van der Waals-coupled two-dimensional materials offer a platform for engineering electronic band structure and fostering emergent
quantum states. Misfit layered compounds provide naturally assembled superlattices, in which crystallographically distinct two-dimensional subsystems coexist within a single bulk
crystal. Therein, the presence of both commensurate and incommensurate
lattice directions introduces structural modulation, which might alter real-space electronic orders such as superconductivity. Here, we investigate the atomic-scale surface structure and local superconductivity of the misfit layered compound (SnS)1.17NbS2 using structural characterization, electrical transport, and low-temperature scanning tunnelling microscopy (STM). While X-ray diffraction and cross-sectional scanning transmission electron microscopy confirm the bulk misfit structure, STM resolves large SnS- and NbS2-terminated terraces, where the surface atomic registry differs from the bulk misfit structure. Furthermore, tunnelling spectroscopy at 2 K reveals nearly identical superconducting gaps on both terminations, and the spatial variation of the gap shows no visible modulation that corresponds to the misfit
lattice. These results establish an atomic-scale picture of the surface structure and superconducting states of this naturally assembled misfit
heterostructure.
Read the full abstract ▾
All authors Dongming Zhao, Shamarita Deb, Lichen Ji, Souvik Das, Paul Simon, Raúl Cardoso-Gil, Steffen Wirth, Walter Schnelle, Sushmita Chandra, Jian-Feng Ge, Claudia Felser
Algebraic Geometry
Moduli descriptions of v-sheaf local models
João Lourenço, Timo Richarz, Eva Viehmann et al. (+1 more)
We propose a uniform moduli-theoretic description of the v-sheaf local model $\mathbb{M}_{\mathcal G,\mu}$ for an arbitrary parahoric group scheme $\mathcal G$ and an arbitrary dominant cocharacter $\mu$. Our approach is Tannakian and builds on a new theory of parahoric induced modules. When $\mu$ is minuscule, this framework explains the linear-algebraic conditions arising in the theory of local models of Shimura varieties.
Read the full abstract ▾
All authors João Lourenço, Timo Richarz, Eva Viehmann, Torsten Wedhorn
Computer Vision
LoopREG: Representation Supervision Enables Test-Time Recurrent Depth in Diffusion Transformers
Zipeng Wang, Anping Xiong, Jing Zhang et al. (+17 more)
Reusing
Transformer blocks offers a way to increase
diffusion-model computation at test time, but naive recurrence can degrade generation beyond its training depth. We introduce LoopREG, a representation-supervised recurrent
diffusion Transformer that repeatedly applies a shared block at the state supervised by a pretrained visual encoder.
Alignment is applied after the final recurrent visit during training; inference extends the same learned operator to additional visits. On ImageNet 256 x 256, LoopREG reduces SiT-B/2 FID from the local REG baseline's 15.08 to 4.91 at 400K training steps, a 67.41% reduction. Within a checkpoint trained with two visits, one extra inference visit improves FID from 14.71 to 5.34 with unchanged weights. The benefit persists on SiT-L/2 and SiT-XL/2. Controlled interventions show that useful recurrence follows the location of representation supervision, while REPA
alignment produces a smaller extrapolation benefit. These results make representation-supervised recurrence a practical route to improving
diffusion generation through additional inference depth.
Read the full abstract ▾
All authors Zipeng Wang, Anping Xiong, Jing Zhang, Xiaoya Zhu, Yi Jiang, Yanbing Xiao, Zige Zeng, Zezhou Chen, Xiang Liu, Huan Hu, Ping Chen, Kai Wang, Zhaoxiang Liu, Shiguo Lian, Ye LV, Fang Zhao, Huanlin Gao, Zhengrui Huang, Bo Shen, Yuren You
Complex Variables
A converse to Littlewood-type theorem for random Dirichlet series
Jiale Chen, Min He and Maofa Wang
This article investigates random Dirichlet series within the framework of Hardy and mixed norm spaces of Dirichlet series. We completely characterize when the random symbol space of the Hardy space $\mathscr{H}^p$ coincides with $\mathscr{H}^2$. In the setting of mixed norm spaces $\mathscr{H}^{p,q}_{\alpha}$, the corresponding characterizations for the random symbol space of $\mathscr{H}^{p,q}_{\alpha}$ coinciding with $\mathscr{H}^{2,q}_{\alpha}$ are given when $q\geq4$, and some sufficient and necessary conditions are established when $q<4$. We also obtain some integrability results when Dirichlet series belong to the corresponding random symbol spaces.
Read the full abstract ▾
Algebraic Geometry
Geometric and homological interpretations of tensor flattenings
Matěj Doležálek, Jakub Jagiełła, Piotr Oszer et al. (+2 more)
We relate several nonlinear flattenings of structure tensors of finite
algebras and modules to their algebraic and homological invariants. Since smoothability of
algebras is equivalent to their structure tensors having minimal border rank, this shows equivalence of several tensor-theoretic nonsmoothability certificates to known classical ones. Particularly, we prove two
conjectures of Doležálek and Michałek relating the tangency flattening and the $3\times3$ minor flattening to dimensions of tangent spaces to the
Hilbert scheme of points. In the process, we generalize the tangency flattening to a mixed regime with two tensor inputs, relating this to dimensions of $\operatorname{Ext}$ modules, and introduce the dual notion of the cotangency flattening, related to dimensions of $\operatorname{Tor}$ modules. In the case of
algebras, we also realize the kernel of the dualized tangency flattening within the setting of Harrison cohomology, making the connection to the tangent space of the
Hilbert scheme explicit.
Read the full abstract ▾
All authors Matěj Doležálek, Jakub Jagiełła, Piotr Oszer, Tim Seynnaeve, Colin Stastny
Machine Learning
Searching for What Matters: Open-Ended Discovery of Multiple Hypotheses with Agentic Data Exploration
Sanchit Kabra, Nurendra Choudhary, Jiangshu Du et al. (+3 more)
Discovering what drives an outcome from data is central to science and often requires deciding what is worth investigating before a specific hypothesis is known. Yet most existing hypothesis generation settings substantially narrow this search by providing a research question, target relationship, or curated context. We introduce open-ended multiple hypothesis discovery, where a system receives only a large tabular
dataset and an outcome and must discover multiple data-supported hypotheses without knowing in advance the relevant variables, functional forms, conditions, or number of underlying mechanisms. To study this problem, we introduce a
benchmark built from structural causal models grounded in real-world data schemas, containing 351 planted hypotheses across 15 tasks and five domains, with 53–65 variables and 1M–10M rows per task. The causal model is hidden from the discovery method and retained only by the evaluator, allowing proposed hypotheses to be tested through interventions and separating observational discovery from causal validation. We further propose HypAgent, which treats the
dataset as an interactive environment and performs iterative search by proposing candidate directions, triaging and refining promising hypotheses, and removing explained signal so that each accepted discovery redirects subsequent search. HypAgent achieves 0.47 precision, 0.51 recall, and 0.49 F1, compared with 0.35 F1 for the strongest baseline. Our analysis shows that many failures occur after a relevant variable has already been surfaced, suggesting that correctly characterizing a mechanism's functional form and conditions, rather than merely identifying relevant variables, remains a key challenge in open-ended discovery.
Read the full abstract ▾
All authors Sanchit Kabra, Nurendra Choudhary, Jiangshu Du, Eddie Huang, Danai Koutra, Chandan K Reddy
Robotics
Information-based Unified Design Framework for NMR Magnetic Tracking System
Seçkin Eren Yetim and Mehmet Efe Tiryaki
Nuclear
magnetic resonance (NMR)
magnetic tracking offers a promising alternative to
magnetic resonance imaging (MRI)-based localization for medical robotic guidance. However, scalar field measurements and field-aligned marker moments constrain position and orientation tracking, requiring careful configuration design. We present a unified information-based framework that uses local covariance analysis to evaluate receiver geometry, marker separation, and external source configuration in on/off-board NMR tracking. Numerical evaluation shows that planar-array precision depends jointly on receiver count and geometry, with triangular
lattice arrays achieving the lowest best-case position uncertainty among the evaluated layouts for 9–20 receivers. Two spatially separated, field-aligned markers enable device-axis tracking, with predicted angular uncertainty below $2^\circ$ at the evaluated workspace center for suitable marker separations. We further establish point-dipole source–receiver duality, unifying the analysis of on/off-board NMR tracking. These
Read the full abstract ▾
Software Engineering
Mobile App Analysis in the New Era: Challenges and the Road Ahead
Jiale Wu, Tianming Liu, Yanjie Zhao et al. (+1 more)
Mobile applications (apps) have grown into complex, multi-platform systems that support diverse services and operate within varying regulatory and distribution contexts. Traditional analysis techniques have established a strong foundation for ensuring quality, security, and compliance, but face increasing challenges as mobile apps evolve. This article presents a three-layer perspective to examine these challenges. The Artifact Layer highlights how heterogeneous code bases and protection mechanisms reduce the analyzability of packaged apps. The Runtime Layer addresses growing execution complexity, from multi-modal interactions and super-app architectures to AI-driven autonomous operations. The Ecosystem Layer considers broader contexts, including emerging platforms, fragmented distribution channels, and region-specific regulatory requirements. New characteristics across these layers interact and compound, creating difficulties that existing approaches are not designed to handle. For each layer, we identify key challenges and potential research directions, including framework-aware analysis, adaptive testing, and policy-aware compliance verification. We also discuss how large
language models bring new analytical capabilities to the field, while their integration into apps raises new security and
privacy concerns. This article aims to clarify where established methods fall short and highlight directions for future research in this evolving field.
Read the full abstract ▾
All authors Jiale Wu, Tianming Liu, Yanjie Zhao, Haoyu Wang
Information Theory
The Covering Radius of the Reed–Muller Code RM(1,9) is 242
Qichun Wang, Yueying Lou and Zifan Wang
What is the covering radius of the Reed–Muller code RM(1,9) ? This is a long-standing open problem in cryptography and coding theory. In this paper, we resolve this problem, and prove that the covering radius of RM(1,9) is 242. We exclude nonlinearity 244 by obtaining a cubic Boolean function achieving a sharp Walsh $\ell^1$ bound. Its equality cases reduce either to an impossible cubic hyperplane restriction or to two homogeneous cubic forms in the Brier–Langevin classification. To exclude nonlinearity 243, we deduce the inequality ${\Vert W_g\Vert }_{1}+2|W_g(0)|\le 10368$, and then reduce either to a spectrum bounded by 24 or to the same expression to be at least 10496.
Read the full abstract ▾
Robotics
LAG-IO: Multi-IMU State Estimation for Legged Robots Using Factor Graphs
Yicheng Zhou, Longwen Chen, Qinghua Yu et al. (+3 more)
Robust proprioceptive state estimation is crucial for legged
robots, particularly in challenging environments where exteroceptive sensors are unreliable. While traditional leg odometry typically relies on a single body-mounted inertial measurement unit (IMU) fused with leg kinematics via an extended Kalman filter (EKF), such methods are vulnerable to linearization errors and hard-contact assumptions, leading to severe drift during slippage or dynamic locomotion.In this paper, we propose LAG-IO, a factor graph-based multi-IMU state estimation framework that tightly couples measurements from a main body IMU, multiple leg-mounted IMUs, and joint encoders within a sliding-window optimization. To handle non-stationary foot-ground interactions, we introduce a continuous probabilistic contact detection module based on kernel density estimation (KDE), which adaptively scales the covariance of zero-velocity update (ZUPT) factors. Furthermore, kinematic relative position factors are formulated to structurally anchor the independent leg inertial states to the
robot's main body. Experimental results on a quadruped
robot across various unstructured terrains demonstrate that, compared to existing baselines, LAG-IO achieves higher planar trajectory accuracy, reduces unobservable vertical divergence, and provides smoother trajectory transitions during unstable contacts, while maintaining competitive attitude observability.
Read the full abstract ▾
All authors Yicheng Zhou, Longwen Chen, Qinghua Yu, Junhao Xiao, Huimin Lu, Xieyuanli Chen
Artificial Intelligence
ECMR: Evidence-Aware Deliberation Control for Multi-Agent LLM Reasoning
Yuan Liu, Haoyue Peng, Fanliang Hu et al. (+2 more)
Multi-
agent Large
Language Model (LLM) systems typically commit to a deliberation protocol before inference. A fixed protocol cannot allocate computation differently to resolved and unresolved instances, and it may mistake repeated support for independent confirmation. We study this failure as semantic false consensus: a confident group answer supported by narrow, duplicated, or correlated rationales. We introduce ECMR, an evidence-aware controller over frozen role-conditioned LLMs. At each step, ECMR summarizes answer concentration, support diversity and independence, contradictions, role behavior, verification history, and remaining budget, then selects whether to propose, challenge, verify, arbitrate, diversify, correct a role, or stop. We formulate this decision as finite-horizon budgeted control and derive sufficient conditions under which state conditioning, risk-triggered verification, and role correction improve cost-adjusted utility. Across a declared held-out evaluation of 9,000 records from MedQA, CaseHOLD, and FinQA, ECMR-Full increases pooled accuracy from 62.1% to 84.9%, reduces false-consensus rate from 40.7% to 17.0%, and raises evidence-support rate from 58.0% to 88.7%. State shuffling loses 11.7 accuracy points with only 0.7 fewer rounds, while harder traces shift from early stopping toward arbitration. ECMR-Full uses the largest
token budget, yielding the strongest absolute reliability rather than uniformly superior
token efficiency.
Read the full abstract ▾
All authors Yuan Liu, Haoyue Peng, Fanliang Hu, Lei Huang, Wenli Cai
Computation and Language
Prefilling-dLLM v2: Learned Chunk Routing for Long-Context Diffusion Language Models
Qi Han, Jing Xiong, Ziyuan Zhang et al. (+1 more)
Diffusion language models apply bidirectional
attention over the whole context at every denoising step, so a longer context makes each step substantially more expensive. Sparsifying only at inference leaves the pretrained weights unchanged, so the model is never trained under the sparse pattern it runs with. We propose Prefilling-dLLM v2, which continues training a
diffusion language model so that it routes with its native mask
token. One mask is appended at the end of each chunk, and a learned role offset marks it as a routing slot. On every fourth layer, a query keeps a bidirectional local window and the top-K chunks scored from that mask; the other layers use a bidirectional sliding window. Starting from Dream-v0-Base-7B extended to 16K with YaRN, we compare Prefilling-dLLM v2 with bidirectional NSA, DSA, and SWA under the same layout, key budget, and dense initialization, using the same Fast-dLLM decoder. With the YaRN factor scaled to the evaluation length, Prefilling-dLLM v2 leads the other sparse methods on all three RULER tasks from 16K to 128K. Its mean accuracy at 128K is 7.4 points above NSA, the next sparse model, while DSA and SWA remain near zero. At the same length, decoding is 172x faster than with full dense
attention.
Read the full abstract ▾
All authors Qi Han, Jing Xiong, Ziyuan Zhang, Ngai Wong
Machine Learning
RT-J: Large-Scale Pretraining of Relational Transformers for Context-Efficient Predictions
Rishabh Ranjan, Vignesh Kothapalli, Harshvardhan Agarwal et al. (+5 more)
Foundation models for tables and relational databases are increasingly used for predictive tasks across various scientific and industrial domains. However, adapting existing models to unseen
datasets requires tens of thousands of examples. Here we introduce RT-J, an 86M-parameter relational foundation model (RFM) that makes accurate predictions even with hundreds of in-context examples or fewer. To achieve this, we first assemble the Join, a large collection of 639 web-crawled databases with 6,023 LLM-generated forecasting tasks. We pretrain with long contexts of up to 8,192 database
cells, using in-context examples from a novel lightweight random-walk retriever. Finally, we leverage context tuning and ensembling to scale up test-time compute. On in-context regression tasks from RelBench, RT-J requires 10x fewer examples on average to match strong feature engineering pipelines with LLM-
agent or RDBLearn featurizers and TabICLv2 or LightGBM predictors. Overall, our work pushes the frontier for context-efficient predictions on relational data and highlights the importance of real-world semantics, context engineering and large-scale pretraining for RFMs.
Read the full abstract ▾
All authors Rishabh Ranjan, Vignesh Kothapalli, Harshvardhan Agarwal, Charilaos Kanatsoulis, Roshan Upendra, Tom Palczewski, Carlos Guestrin, Jure Leskovec
Computer Vision
MicroCLIP: Connecting Text and Microscopy Images via Contrastive Language–Image Pretraining
Yanran Ruan, Zhixiang Wei, Shiyao Hong et al. (+5 more)
Understanding fluorescence microscopy images is fundamental to automating biological research. However, existing methods typically learn predefined concepts from closed-set data, limiting transfer across tasks and biological contexts. In this work, we introduce MicroCLIP, a
vision-
language model that learns generalizable cellular and subcellular representations from natural
language supervision for open-set image classification and
retrieval. To pretrain MicroCLIP, we construct MicroPool-2M, a
dataset of approximately 2.5 million fluorescence microscopy image-text pairs spanning diverse cellular structures,
cell lines, and imaging conditions. Aligning microscopy images with biological concepts in natural
language is challenging because most images lack descriptive captions and biological knowledge annotations. We therefore progressively establish image-text associations at three levels to provide MicroCLIP with richer semantic supervision: (1) category-level
alignment links images to standardized labels defined by subcellular structure-
cell-line combinations; (2) knowledge-level
alignment incorporates biological priors derived from these labels to capture relationships across species and
cell types; and (3) detail-level
alignment uses a
multimodal large
language model to recaption and verify image-text pairs at scale, emphasizing fine-grained visual characteristics. Extensive experiments demonstrate that MicroCLIP achieves leading performance in zero-shot classification and text-image
retrieval on microscopy images. It retrieves images with specific cellular characteristics from thousands of candidates and generalizes to unseen biological concepts without additional supervision. These results highlight MicroCLIP's potential as a foundation model for fluorescence microscopy.
Read the full abstract ▾
All authors Yanran Ruan, Zhixiang Wei, Shiyao Hong, Xiaoxiao Ma, Yuxuan Gu, Liming Ye, Huaian Chen, Yi Jin
Number Theory
Fine structure of even Galois deformation rings
Peter Vang Uttenthal
For every even Galois representation in a family of
prime conductor $\ell$, we prove that if $k \in \mathbb{Z}_{\geq 1}$ is maximal such that \[\ell \equiv 1 \bmod 3^k\] then the even mod $3$ deformation ring $R_S \otimes_{\mathbb{Z}_3} \mathbb{F}_3$ of level $S=\{\ell,3\}$ does not have a quotient isomorphic to $\mathbb{F}_3[T]/(T^{2+3^{k-1}})$. As a corollary, we deduce that the global even deformation rings $R_S$ are finite and flat in the category of $\mathbb{Z}_3$-modules.
Read the full abstract ▾
Machine Learning
TStruct: Learning Shared Temporal Structures for Long-Term Time-Series Forecasting
Yangze Li, Zhengyang Zhou, Qihe Huang et al. (+3 more)
Long-term time series forecasting requires temporal dependency modeling that remains stable and generalizable under temporal distribution shifts. However, real-world time series are often noisy and non-stationary, making deep learning models prone to fitting incidental temporal interactions in the training data. A key question is whether there exist temporal relationship structures that recur across different temporal positions and variables. We analyze temporal relationship
matrices on commonly used time series
datasets and find that samples from different temporal positions and different variables exhibit shared dominant temporal relationship structures. This indicates that temporal relationship structures contain patterns that recur stably across samples and variables. To learn such shared structures, we propose Temporal Structure Network (TStruct), a simple yet effective network for directly learning shared and generalizable temporal dependencies. Specifically, TStruct directly parameterizes a set of shared temporal dependency
matrices and generates normalized combination weights using dynamic sample information and static variable priors. These weights represent a soft allocation of each sample-variable pair over shared temporal dependency structures, enabling the model to generate adaptive temporal dependency
matrices under shared structural constraints, rather than fitting incidental interactions in a fully unconstrained manner. Notably, this simple structure-aware design consistently outperforms several strong baselines across all
datasets, achieving superior accuracy while maintaining high computational efficiency.
Read the full abstract ▾
All authors Yangze Li, Zhengyang Zhou, Qihe Huang, Pengkun Wang, Binwu Wang, Yang Wang