arXiv subjects

Bo Yang

Publications and source records attributed to Bo Yang.

At least 19 recordsLinked to original sources

arXiv2026

Guided-RLVR: Guided Exploration in Reinforcement Learning with Verifiable Rewards

Reinforcement learning with verifiable rewards (RLVR) has shown strong empirical gains in improving the reasoning ability of large language models (LLMs) on tasks such as mathematics and programming. However, recent studies have shown that RLVR rollouts often collapse onto a narrow set of reasoning trajectories, limiting the explored reasoning space. Motivated by findings that thinking tokens at critical steps serve as critical structural elements in LLM reasoning pathways, we propose Guided-RLVR, a novel approach that augments RLVR with a lightweight controller. The controller strategically injects semantically meaningful steering tokens into the sampling distribution, encouraging the model to approach problems from multiple reasoning perspectives and thereby expanding the exploration space. Experiments on mathematical reasoning and code generation benchmarks show that Guided-RLVR consistently outperforms existing RLVR methods while producing more diverse reasoning trajectories.

cs.AI↗

arXiv2026

A switching proximal gradient method for constrained difference-of-convex optimization problems

We study difference-of-convex optimization problems with several composite inequality constraints and develop a majorization-guided switching proximal gradient (MGSPG) method. Specifically, at each feasible outer point, quadratic majorization and linearization yield a strongly convex composite surrogate problem. To solve this surrogate approximately, we employ a switching proximal gradient (SPG) method, which takes a proximal gradient step for the objective when the current point is nearly feasible and for a most violated constraint otherwise. This construction requires neither projection onto the joint surrogate feasible set nor multiplier computation. By establishing a common invariant ball and using an increasing weight argument, we obtain an $O(τ^{-1})$ inner complexity bound for SPG. Building on this bound, we show that, under boundedness of the feasible level set and a uniform Slater condition for the surrogates, MGSPG finds an $\varepsilon$-KKT point in $O(\varepsilon^{-2})$ outer iterations and $O(\varepsilon^{-4})$ selected component proximal gradient updates. Numerical experiments on SCAD-constrained and groupwise-constrained regression problems demonstrate the viability and computational efficiency of our method compared with existing optimization algorithms.

math.OC↗

arXiv2026

Neural Compression of Long ADMM Trajectory for Multiparametric Quadratic Program

Solving large-scale multiparametric quadratic programs (mpQPs) in real time often requires thousands of iterative optimization updates, creating a major computational bottleneck in model predictive control and learning-enabled systems. This paper proposes TQPNet, a self-supervised neural compression framework that learns to compress long ADMM optimization trajectories into a compact neural warm start followed by a few refinement iterations. We derive a low-dimensional ADMM scheme operating on the primal-dual variables $(x,β)$ and show that its iteration admits an equivalent ReLU-layer representation, termed ADMM(ReLU). Building on this insight, TQPNet combines a multilayer perceptron with ReLU activations (MLP(ReLU)) and a small number of unrolled ADMM(ReLU) refinement layers. The network is trained offline using an optimality-condition-informed residual loss, eliminating the need for solver-generated labels. The proposed residual-based framework enables both self-supervised training and online solution certification through adaptive ADMM(ReLU) refinement. Experiments on large-scale mpQPs and real-time MPC applications demonstrate that TQPNet achieves high accuracy, strong generalization, and substantially faster inference than conventional optimization-based methods.

math.OC↗

arXiv2026

PhyMo: Learning Physical Dynamics with Accurate and Continuous Motion from Multi-View Videos

Recovering dynamic 3D scene geometry, appearance, and the physical motion states that govern scene evolution from multi-view videos is important yet challenging. A key difficulty is that appearance can be deceptive. Under rendering supervision alone, a model may fit the observed images by changing color or opacity to compensate for inaccurate deformation, rather than recovering the true evolving geometry and motion. This issue is especially severe in scenes with fast or locally complex motion, where inaccurate trajectories can still produce plausible images over the observed frames, but lead to unreliable reconstruction and discontinuous future prediction. In this paper, we propose PhyMo, a framework for learning evolving dynamics from multi-view videos. We introduce a Motion-Aware Trajectory Representation that drives the model to recover accurate motion from deformation, rather than relying on appearance shortcuts to explain observed motion. On top of this, we propose a Continuous Motion Evolution module with bidirectional kinematic continuity constraint to promote temporally consistent evolution of displacement, velocity, and acceleration. Together, these designs yield geometrically faithful interpolation and continuous motion extrapolation. Experiments on four dynamic scene benchmarks demonstrate state-of-the-art performance.

cs.CV↗

arXiv2026

A Calculus of Inheritance

Just as the $λ$-calculus is a pure functional programming language built from three primitives (abstraction, application, variable), we present inheritance-calculus, a pure declarative programming language built from three primitives (mixin, definition, reference). We introduce multi-target $\mathtt{this}$ to solve the problem that inheritance conflicts with modularity, and thereby unify object-oriented and configuration constructs: modules, classes, objects, methods, fields, and locals become a single mixin abstraction, and inheritance becomes simply set union, the deep merge of configuration languages. Consequently, composition is commutative, idempotent, and associative, structurally eliminating the multiple-inheritance linearization problem; the inheritance relation has substitutability; every path of every program can be inherited and extended; every subtree can be split into parts that, inherited together, behave as the whole; and programs have no side effects and cover every input. We call these properties modularity by construction. The calculus has no functions, yet it macro-expresses the lazy $λ$-calculus and is more macro-expressive than it in Felleisen's sense. Its semantics is first-order, denotational, and evaluated by tabling, even for cyclic inheritance hierarchies. Inheritance-calculus is distilled from MIXINv2, an implementation in which the same code acts as different function colors and ordinary arithmetic agrees with the relational semantics of logic programming.

cs.PL↗

arXiv2026

Capillary John ellipsoid theorem with applications to capillary curvature problems

In this paper, we apply a capillary John ellipsoid theorem for capillary convex bodies in the Euclidean half-space $\overline{\mathbb{R}^{n+1}_{+}}$. This theorem yields a non-collapsing estimate for capillary hypersurfaces, which provides a new approach to obtaining $C^{0}$ estimates for solutions to some capillary curvature problems (including the capillary $L_{p}$ Christoffel-Minkowski problem and the capillary $L_{p}$ curvature problem), based on the corresponding gradient estimates. As an application, we study the capillary $L_{p}$ dual Minkowski problem. A gradient estimate, together with the non-collapsing estimate and a $C^2$ estimate, yields existence for every $1 1$, with uniqueness up to dilation. We further obtain existence and uniqueness for $p>q$ without an upper restriction on $q$.

math.AP↗

arXiv2026

Task-Oriented Sensing and Covert Transmissions for Collaborative Multi-AUV Systems

In underwater covert cooperative missions, autonomous underwater vehicles (AUVs) often cannot rely on active sonar to continuously obtain complete information, since active sensing and frequent communications increase exposure risk. As a result, AUVs primarily rely on passive observation, which yields incomplete local perception and limited task efficiency. Although underwater acoustic communications can mitigate this limitation through information sharing, they are simultaneously constrained by long delays, severe interference, low reliability, and the risk of covert exposure. Existing communications-oriented multi-agent reinforcement learning (MARL) studies often model communication as an ideal information flow, whereas traditional communication optimization primarily focuses on link-level performance. However, both are insufficient to characterize the actual contribution of perceptual information to cooperative tasks under realistic covert physical communication conditions. This paper proposes a Sensed-information Value Realization Multi-Agent Reinforcement Learning (SVR-MARL) framework that leverages practical information to characterize information utility for cooperative tasks and learns distributed cooperative policies under realistic communication and covert constraints. Through a case study of covert multi-AUV cooperative localization and tracking, the potential of the proposed framework to improve collaborative task efficiency while reducing unnecessary communication and exposure risks is demonstrated.

cs.LG↗

arXiv2026

Nash-MOPD: Optimizing Task Trade-offs in Multi-Teacher On-Policy Distillation

Multi-teacher on-policy distillation (MOPD) aggregates the capabilities of multiple domain-expert teachers into a single student model. However, all domains in MOPD share the student's finite budget of optimization steps and rollouts. Fixed uses constant domain sampling ratios based on corpus sizes; by contrast, D$^3$-MOPD adjusts the ratios using the teacher-student reverse-KL divergence and its recent decrease. However, favoring large recent decreases can direct more training toward domains that already improve quickly, while domains that improve more slowly may receive too few examples. To address inflexible allocation and the risk of giving insufficient training to domains that improve more slowly, we propose Nash-MOPD, an adaptive domain scheduler based on a local minimax game using existing reverse-KL statistics. The equilibrium maximizes the smallest predicted proportional reduction under the local model and gives more prompts to domains that need more examples for comparable improvement. Consequently, Nash-MOPD responds to changing learning progress and supports domains that improve more slowly without extra teacher queries or rollouts. At step 100, Nash-MOPD exceeds both baselines' final mathematics and code checkpoint scores on DeepSeek-R1-Distill-Qwen-1.5B and reaches Fixed's on Qwen3-8B-Base. Furthermore, it leads eight of nine reported metrics and is best or tied best on all six mathematics and code accuracy metrics, respectively.

cs.LG↗

arXiv2026

Cyclic Graphs and Memoization in Pure $λ$-Calculus

Purely functional programming makes immutability and non-strict evaluation its defaults; we push further, making interning and memoization the defaults of computation. In prior operational semantics for the pure $λ$-calculus, representing and transforming cyclic and infinite data requires an added recursion construct, a $\texttt{letrec}$, a $μ$-binder, or a built-in $Y$ for graph reduction, and sharing the repeated work of a memoized or dynamic-programming function requires an impure cache. We show that no extension is needed. We apply tabling, the standard method for solving a least-fixpoint equation, to weak-head reduction; this defines a new operational semantics for the pure $λ$-calculus that keeps each term's standard lazy meaning. A term that reaches finitely many distinct states, each solved in finitely many steps, comes out as a finite graph, possibly cyclic; the calculus stays pure, and the graph is faithful and independent of reduction order. We implemented this operational semantics as a $λ$-calculus interpreter. It does dynamic programming automatically, sharing repeated subproblems with no memoization table. It creates and transforms cyclic graphs with no added recursion construct. And it decides in finite time that some terms have no weak head normal form, returning $\bot$ for $Ω$. What the evaluator returns is a graph, so the $λ$-calculus becomes a DSL for graph computation: the memo table of dynamic programming, the transposition table of game search, and the derived-fact table of Datalog reachability and points-to analysis are all tabling on state identity, and none of them is written by hand. Compilation is one more such problem: we write a bootstrap compiler that compiles its own source, all as a pure $λ$-term.

cs.PL↗

arXiv2026

Kähler geometry on total spaces of vector bundles over elliptic curves

We study function theory and Kähler geometry on total spaces of vector bundles over an elliptic curve. Our main result characterizes the vector bundles whose total spaces admit complete Kähler metrics with nonnegative bisectional curvature ($BI \geq 0$ for short). We also obtain explicit expressions for all complete Kähler metrics with $BI \geq 0$ on the total space of a line bundle over a complex torus. These results are related to a general conjecture concerning the structure of complete Kähler manifolds with $BI \geq 0$. For rank-two vector bundles of degree zero over elliptic curves, we study holomorphic functions and automorphisms on the corresponding total spaces, and construct complete Gauduchon Hermitian metrics with vanishing Chern-Ricci curvature. These metrics are natural in the sense that the associated spaces of holomorphic functions of polynomial growth coincide with the ``polynomials'' on these total spaces.

math.DG↗

arXiv2026

Redundant or Necessary? A Benchmark for Detecting Redundant Steps in Agent Trajectories

Computation cost is a critical challenge for LLM agent systems, as their long-horizon trajectories continuously accumulate computational overhead. While existing studies focus on improving the overall efficiency of agent systems, the efficiency of individual steps remains largely unexplored. In this paper, we propose and formulate a new research problem: \textbf{redundant step detection}, aiming to detect these steps that consume substantial computational resources while contributing little or nothing to task completion. To support this initiative, we introduce \textbf{RedundancyBench}, a new benchmark that contains 1,920 trajectories collected from 6 different LLMs with 3,613 step-level annotations. Using RedundancyBench, we develop and evaluate 6 frontier LLMs and Jev as redundant step detectors, summarizing their strengths and limitations. Our experiments reveal that the redundant step detection is still challenging even for SOTA models, such as Claude-Opus-5, particularly without prior information. Code and dataset in this paper are both available in https://anonymous.4open.science/r/RedundancyBench-E0CB.

cs.AI↗

arXiv2026

Holomorphic functions on a class of complete Hermitian manifolds

In this paper, we prove a gradient estimate for holomorphic functions on complete Hermitian manifolds with nonnegative second Ricci curvature. As an application, we establish several Liouville-type theorems for holomorphic functions and mappings between complete Hermitian manifolds under suitable curvature assumptions. In particular, on a complete Hermitian manifold with nonnegative second Ricci curvature and Chern torsion of linear growth, any bounded holomorphic function must be constant. Moreover, we construct examples to show that both the gradient estimate and this Liouville theorem are sharp in a suitable sense. In the second part of the paper, we study holomorphic functions on simply connected complete Hermitian manifolds with vanishing Chern curvature and torsion norm of sublinear growth. Combining the gradient estimate with methods from sub-Riemannian geometry, we obtain a complete characterization of holomorphic functions of polynomial growth on such manifolds.

math.DG↗

arXiv2026

FiberTune: Preserving Action-Fiber Visual Residuals in Vision-Language-Action Fine-Tuning

Action-supervised fine-tuning of vision-language-action (VLA) policies can compress visual structure that is weakly constrained by the action objective. We formalize this as residual visual collapse along local action fibers and propose FiberTune, a training-time objective that preserves teacher-structured visual residuals without adding inference-time overhead. FiberTune uses an online action probe to filter action-predictive directions from intermediate visual-token representations. It preserves the residuals through centered directional alignment with a frozen visual teacher and effective-rank regularization. Under identical training conditions, FiberTune improves over task-loss-only fine-tuning in every one of six simulation settings spanning two benchmarks and two architectures ($π_{0.5}$ and OpenVLA-OFT), as well as on physical SO-101 pick-and-place; representative gains include +10.7 pp SR(5) on long-horizon CALVIN ABC$\rightarrow$D and physical SO-101 task success rising from 72.7% to 78.1%. Representation analyses support the preservation of visual information in the residuals.

cs.CV↗

arXiv2026

MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement

Reinforcement learning (RL) is the central training paradigm for advancing large foundation models towards self-improvement. This report introduces the MiMo-V2.6 series, an omni-modal family that pushes the frontier of model intelligence by scaling RL compute. Prior to RL, we conduct mid-training on a broad multimodal corpus to provide ample exploration space, and build a solid infrastructure on the pretrained hybrid-SWA architecture to support subsequent scale-up. We scale RL compute along three dimensions: (1) larger batches and higher throughput, with an asynchronous training that consumes 1,568 samples and 2.7-3.7B tokens per step at context lengths of up to 1M; (2) more diverse and complex environments, spanning code, general, visual, and cyber domains under a mixture of agent harnesses; and (3) more grader compute, via groupwise agentic grading that yields more accurate reward signals for long-horizon tasks and steers the model towards shorter, more token-efficient solutions. To keep training stable at scale, we freeze the MoE router and establish a multi-layer defense against reward hacking. We further build infrastructure for mixed-task agentic RL, including a unified trajectory representation, high-concurrency multi-framework rollout, decoupled control and data planes, and training-inference consistency. We open-source the training dynamics, RL environments, and RL framework to facilitate reproduction and further research on scaled RL and model self-improvement.

cs.CL↗

arXiv2026

The Attribution Blind Spot: Layerwise Trajectory Diagnostics for Source Reliance in Retrieval-Augmented Language Models

A retrieval-augmented model can match a document without relying on it. Controlled knowledge conflicts make source choice observable and let us ask a second question that prediction alone cannot answer: which internal-state properties define useful intervention directions? We study paired hidden-state changes with Latent Trajectory Shift (LTS), a signed projection onto a training-fitted first principal component (PC1), and keep verified training exposure separate from behavioral source choice. Across the evaluated conflicts, state-change magnitude is often the stronger predictor, whereas signed PC1 is the stronger selective controller: equal-norm interventions change source preference while better preserving non-target behavior, and the frozen direction transfers across the tested datasets and aligned model pairs. A same-system OLMo study further combines positive choice and control results with inconclusive exposure detection at the achieved power. The central result is a separation: representations that diagnose what a model will choose need not be the representations that best control that choice.

cs.AI↗

arXiv2026

Visualizing Three-Dimensional Quantum Fluctuations of Ground-State C$_3$H$_4$ via Deep Sub-Angstrom Coulomb Explosion Imaging

Coulomb explosion imaging (CEI) is a powerful tool for determining static geometries of isolated gas-phase molecules. In this work, taking allene and propyne as prototype, we demonstrate sub-angstrom resolution achieved with highly charged ion-induced CEI. Using 41.9 keV/u Xe$^{20+}$ beam, minute variations of CC single, double and triple bond lengths are clearly visualized, indicating that an ultrahigh resolution of 0.1 $\mathring{\mathrm{A}}$ is achieved. This unprecedented resolution not only makes it possible to distinguish CC bonds with different bond orders, but also lays a solid foundation for direct observation of quantum fluctuations inherent to 3D structure of a ground-state molecule. Notably, the coupled collective motion of the two CH$_2$ moieties spaced by a central carbon within allene is unambiguously visualized in our measurements.

physics.chem-ph↗

arXiv2026

Estimation is Not Enough: Carpet-Bombing Detection via Per-Packet Uniformity Testing

Carpet-bombing attacks spread traffic uniformly across one or more destination IP prefixes, keeping every host in the prefix below alarm thresholds while exhausting prefix-level defenses. Existing carpet-bombing detectors run at seconds-to-minutes latency, too slow to respond within the attack window. Sketches support per-packet processing in fixed-width memory, a natural fit for cutting latency, yet no sketch-based detector exists for carpet bombing. Because source addresses can be spoofed, and attacks can be launched through reflection, source-side evidence is structurally unavailable and detection must anchor at the destination side. We present SweepSketch, the first sketch-based detection model for carpet bombing: it anchors at the destination, keeps no source state, and compresses a tagged self-cleaning T-HLL primitive, per-packet CUSUM decisions, and dual-EWMA change gates into 44-byte fixed-width buckets deployable on the Tofino2 programmable switch. Its design is supported by six theorems, including verifiable detection lower bounds. Under the same memory budget, SweepSketch leads all 10 baselines across the sketch, entropy, and sequential-testing classes in F1 (0.991). Its median alarm latency is 186--627 ms, and it has structural immunity to source spoofing. On 30 real/synthetic multi-prefix samples held out from parameter design, it detects all 140 victim prefixes.

cs.CR↗

arXiv2026

pikit: A Composable Toolkit for Indirect Prompt Injection Research and Evaluation

Indirect prompt injection embeds malicious instructions within external content retrieved by LLM-based agents, altering target behavior without user authorization. We introduce pikit, a research toolkit designed to systematically evaluate these threats across three core dimensions: attacks (13 methods), channels (16 carriers across text and file modes), and defenses (9 prevention strategies and 3 offline detection baselines). Built on a decorator-based registry, pikit enables seamless extension of custom components without modifying core code, while a unified craft() API composes arbitrary attacks and channels in a single call. We evaluated the toolkit on the pi coding agent powered by an anonymized LLM in a production-like environment. Benchmarking 9 prevention strategies against high-risk attacks yields a 71.8\% relative reduction in attack success rate, with few\_shot\_warning and instruction\_hierarchy providing the strongest protection. Offline detection baselines achieve perfect precision but low recall, demonstrating that heuristic detectors complement rather than replace prompt-level defenses. To ensure reproducibility, each run automatically logs full prompts, agent event traces, session transcripts, and verdict records. Our code is available at https://github.com/Tencent/AI-Infra-Guard/tree/main/Research/pikit.

cs.CR↗