arXiv subjects

Weikang Li

Publications and source records attributed to Weikang Li.

At least 19 recordsLinked to original sources

arXiv2026

SpatialScout: Can VLM Agents Actively Explore and Map 3D Scenes?

Spatial intelligence requires agents to actively acquire visual evidence and integrate observations into a coherent representation of the three-dimensional world. Yet spatial question answering offers limited evidence of whether models can construct accurate metric maps through autonomous exploration. We introduce SpatialScout, a benchmark for evaluating vision-language model (VLM) agents on active object-level 3D mapping. SpatialScout comprises 80 diverse indoor scenes containing 3,140 target objects. Given an object inventory, RGB observations, camera calibration, and camera poses, agents control a camera to explore each scene and estimate object centers in a shared coordinate frame. The task requires coordinating viewpoint selection, cross-view association, geometric estimation, and map revision under bounded interaction budgets. Evaluation jointly measures localization accuracy and target coverage through object-scale overlap scores that penalize missing predictions. Experiments with frontier VLMs reveal substantial limitations: the strongest evaluated model, Claude Opus 5.5, achieves a mean overlap score of 41.71%, with 47.23% of targets exceeding an intersection-over-union (IoU) threshold of 0.5 and only 3.25% exceeding 0.9. These results reveal a substantial gap between recovering coarse object layouts and constructing complete, precise metric maps. SpatialScout provides a quantitative testbed for developing agents that build reliable 3D spatial representations through active perception. Code is available at https://github.com/JesseZZZZZ/SpatialScout.

cs.CV↗

arXiv2026

Enhancing Bug-Exposing Test Generation via Step-Level Credit Assignment and Self-Distillation from Valuable Failures

Large Language Models have shown strong code generation capabilities, yet their outputs often contain subtle bugs. Automatically repairing such code requires informative tests: inputs that expose faulty behavior, paired with correct expected outputs. Recent RL-based test generation methods apply the same outcome reward to all reasoning steps, which can discourage useful reasoning and reinforce flawed steps. They also fail to effectively learn from trajectories that find bug-triggering inputs but predict incorrect outputs, leaving their verified bug-triggering progress insufficiently exploited. To address these limitations, we propose SCALF, a process-supervised RL approach that combines Step-level Credit Assignment with targeted Learning from valuable Failed trajectories. SCALF derives step-specific advantages from process rewards, allowing each reasoning stage to receive its own learning signal. It further routes trajectories that expose bugs but predict incorrect outputs to on-policy self-distillation, using execution feedback to learn from their valuable bug-triggering progress. Across three benchmarks, SCALF outperforms strong prompting and post-training baselines by up to 2.97, 5.26, and 3.53 percentage points in joint success rate, output correctness rate, and bug-triggering rate, respectively. Our code and datasets are available at https://anonymous.4open.science/r/SCALF-F4D1.

cs.SE↗

arXiv2026

Electrostatic Limits on Quantum Spatial Search

We prove that a quantum walker cannot search a two-dimensional lattice in Grover's $O(\sqrt N)$ time, resolving the long-standing open question posed by Aaronson and Ambainis in 2003. A walker that hops between neighbouring sites of an $L\times L$ lattice, $N=L^2$, needs $Ω(\sqrt{N\log N})$ local steps at least to find a marked vertex, even with unlimited memory and free queries. Hence, Tulsi's algorithm is asymptotically optimal. Our proof rests on a surprising correspondence with classical electrostatics. We treat the marked vertex as a point charge and weight the overlap between the walker's states with and without the mark by the potential drop, which vanishes where queries act. Only local steps change this weighted overlap, at a rate limited by the field energy, so the lower bound is set by the self-potential of the charge. In the plane this self-potential grows logarithmically with system size, like the Coulomb potential, leading to the extra root-logarithmic factor. Several marked vertices turn the charge into a conductor, and the lower bound depends on their arrangement through its capacitance.

quant-ph↗

arXiv2026

A modular quantum computer based on Cornucopia codes

Ultra-high-rate quantum error correction codes store many logical qubits in a single code block with few physical qubits. A fault-tolerant quantum computer, however, also requires a universal set of logical operations on individually addressable logical qubits, and the ancilla overhead of these operations may erode the advantage of the high encoding rate. Here, we propose the Cornucopia architecture, in which a universal set of logical operations is implemented at low overhead for the Cornucopia codes. In particular, we focus on the [[2844,1426,18]] Cornucopia code as an example and introduce a modular quantum computer built from such code blocks connected by bridges. Each block stores 1,426 logical qubits and serves as a universal logical processor. Exploiting the code's translation symmetry, we construct a family of reusable gadgets that individually address 1,422 logical qubits within each block. These gadgets support three universal logical instruction sets: basic, high-throughput and fixed-extractor. Excluding magic-state factories, a module with sufficient workspace to execute any instruction in the basic set requires an average of 4.1 physical qubits per logical qubit, compared with 3.0 for memory alone. The architecture thus preserves the resource advantage of ultra-high-rate quantum memory while enabling universal logical computation, providing a route towards low-overhead fault-tolerant quantum computers.

quant-ph↗

arXiv2026

Reducing Cognitive Overhead in Tool Use via Multi-Small-Agent Reinforcement Learning

Recent advances in multi-agent systems highlight the potential of specialized small agents that collaborate via division of labor. Existing tool-integrated reasoning systems, however, often follow a single-agent paradigm in which one large model interleaves long-horizon reasoning with precise tool operations, leading to cognitive-load interference and unstable coordination. We present MSARL, a Multi-Small-Agent Reinforcement Learning framework that explicitly decouples reasoning from tool use. In MSARL, a Reasoning Agent decomposes problems and plans tool invocations, while multiple Tool Agents specialize in specific external tools, each trained via a combination of imitation learning and reinforcement learning with role-specific rewards. On mathematical problem solving with code execution, MSARL significantly improves reasoning stability and final-answer accuracy over single-agent baselines. Moreover, the architecture generalizes to diverse tool-use tasks, demonstrating that cognitive-role decoupling with small agents is a scalable blueprint for multi-agent AI design.

cs.AI↗

arXiv2026

Efficiently Optimizing the Quantum Value of Bell Inequalities using Batched Gradient Descent

Exploring the world of Bell inequalities requires numerical methods that scale effectively with the number of parties, inputs, and outputs. Such methods can be used to study large-scale Bell inequalities beyond the reach of analytic methods, to design experiments for violating more complicated Bell inequalities, and to enable applications of Bell inequality violation. Such applications include randomness generation and multi-agent coordination with restricted communication. However, current general purpose optimizers such as the see-saw method struggle to handle Bell inequalities with only a dozen inputs and outputs. We introduce an optimizer for the quantum value of a Bell inequality based on batched gradient descent (BGD). More precisely, our BGD optimizer is a differentiable search over feasible quantum strategies in which states and projective measurements are generated from unconstrained variables. The Bell expression is evaluated by a direct tensor contraction without forming the prohibitively large Bell operator. This formulation makes parallel random restarts inexpensive and is thus naturally implemented by a GPU. We evaluate our optimizer against the see-saw method and find a significant speedup for a wide variety of Bell inequality families. Implemented on a GPU, our BGD optimizer can optimize Bell inequalities in our numerical experiments with more than a thousand inputs or outputs within minutes. For these large-scale Bell inequalities, we also efficiently compute the classical value by enumerating possible strategies using a GPU or expressing the optimization as a mixed-integer linear program. Our optimizers can serve as numerical evaluation tools for Bell experiments and multi-agent coordination problems modeled by Bell inequalities with many inputs and outputs, a common feature of real-world scenarios such as high frequency trading and distributed systems.

quant-ph↗

arXiv2026

Proving olympiad geometry theorems on a superconducting quantum processor

Automated theorem proving seeks to use computational systems to prove or disprove mathematical and logical statements [1, 2]. It underpins a wide range of applications, and enhancing theorem-proving capabilities remains a central objective in artificial intelligence [3]. Although recent neuro-symbolic systems have achieved remarkable progress [4-7], their operation is ultimately constrained by classical computational architectures. Quantum computing [8], by contrast, enables information encoding and coherent parallelism beyond classical limits [9-14], raising the possibility of accelerating structured symbolic deduction [15]. Here we report the experimental realization of automated geometry theorem proving on a fully programmable superconducting quantum processor. We develop two complementary quantum proving frameworks. The first implements Wu's algebraic elimination method using quantum pseudo-division, with multivariate polynomials represented in superposition states, enabling quantum algebraic theorem proving. The second implements the full-angle method as backward symbolic reasoning through a hybrid quantum strategy-guided architecture, demonstrating a general route toward quantum symbolic proof search. As illustrative examples, we prove two theorems on a superconducting quantum processor: the perpendicularity of the diagonals of a square and a 1978 International Mathematical Olympiad geometry problem. Our results establish, at the experimental level, automated logical reasoning as a viable task for near-term quantum processors and provide a concrete pathway toward quantum-enhanced symbolic intelligence.

quant-ph↗

arXiv2026

ActTraitBench: Quantifying the Knowledge-Decision Gap in Large Language Models via Human-Grounded Behavioral Validation

While Large Language Models (LLMs) can convincingly simulate personas in explicit self-reports, they often deviate in implicit behavioral decisions, revealing a substantial Knowledge-Decision Gap ($G_{\mathrm{KD}}$). Existing benchmarks struggle to measure this discrepancy due to limited construct validity, multidimensional entanglement, and distributional biases in LLM-based evaluation. To address these issues, we propose ActTraitBench, a human-grounded evaluation framework for measuring personality consistency in LLMs. Grounded in empirical human data, ActTraitBench establishes one-to-one mappings between psychometric facets and behavioral paradigms and applies Distributional Calibration via Quantile Mapping to reduce distributional mismatch between LLM-judge scores and human responses. Experiments on 14 mainstream LLMs reveal substantial knowledge-decision gaps and show that assigned personas are reflected more consistently in self-reports than in behavioral decisions for most evaluated models. To mitigate this gap, we further introduce the Chain of Cognitive Alignment (CoCA), an inference-time intervention that reduces $G_{\mathrm{KD}}$ for 12 of the 13 models with paired results. Code and resources are available at https://github.com/Selina233/ActTraitBench.

cs.CL↗

arXiv2026

Trait-Aware Policy Optimization for Autoregressive Multi-Trait Essay Scoring

Multi-trait essay scoring aims to provide fine-grained evaluation of writing quality across multiple dimensions. However, how to effectively post-train autoregressive scoring models remains underexplored. In this paper, we propose Trait-Aware Policy Optimization (TAPO), a post-training framework tailored to autoregressive multi-trait scoring. Our method decomposes rewards along both the sample and trait dimensions, combining global scoring consistency, trait-level accuracy, format validity, and inter-trait dependency preservation. In addition, we use enhanced prompts throughout training by incorporating original prompt texts and trait descriptions, providing richer semantic information for trait-specific score generation. Experiments across multiple backbone models show that our method consistently improves multi-trait scoring performance over supervised fine-tuning and scalar-reward optimization baselines, demonstrating the effectiveness and transferability of trait-aware post-training for essay scoring.

cs.CL↗

arXiv2026

Quantum error correction at ultra-low overhead

Suppressing errors is the central challenge for useful large-scale quantum computing. While quantum error correction promises a viable solution to this challenge, existing codes typically suffer from trade-offs among encoding efficiency, error threshold, and hardware feasibility. Here, we introduce Cornucopia codes, a family of practical, hardware-efficient quantum low-density parity-check codes that achieve an ultra-high encoding rate exceeding $1/2$ while maintaining a pseudo-threshold exceeding $0.4\%$ under the standard circuit-level noise model. Inspired by recent affine-permutation-based code constructions and the long-range connectivity available in reconfigurable neutral-atom arrays, we adopt a structured code geometry in which the code layout, atom rearrangement, and syndrome-extraction schedule are co-designed. This structure enables nonlocal syndrome measurements through simple, parallel atom rearrangements. A complete syndrome extraction cycle measures all $X$- and $Z$-type checks in parallel with $12$ entangling layers, independent of the code size. The resulting threshold is comparable to those of the surface code and bivariate bicycle codes. In particular, a single code block $[[2844,1426,18]]$ encodes $1{,}426$ distance-$18$ logical qubits, achieving an extrapolated logical error rate of $2.6\times10^{-16}$ ($1.9\times10^{-31}$) per logical qubit per cycle, assuming the physical error rate of $0.1\%$ ($0.01\%$). By comparison, a bivariate bicycle code implementation would require more than $68{,}000$ physical qubits to encode the same number of logical qubits at a comparable logical error rate. These results bring demonstrations of ultra-low-overhead quantum error correction within the reach of near-term quantum processors.

quant-ph↗

arXiv2026

Cluster-State Witnesses of Finite-Speed Hidden Influences

Bell experiments rule out local common-cause explanations of quantum correlations, yet they do not exclude hidden influences that travel faster than light while still having a finite speed in a preferred frame. Multipartite spacetime arrangements turn this possibility into a constraint: two late parties that are outside each other's hidden-influence cones must remain Bell-local once the earlier events are fixed. Here, we formulate this constraint as a projected-polytope separation problem for cluster-state correlations, using only marginal data containing at most one late party. From linear cluster states, we construct a four-qubit witness with the bound $S_4\le 6$ and quantum value $4+2\sqrt2$, and a five-qubit witness with $S_5\le 10$ and quantum value $6+4\sqrt2$. We certify that the exposed faces are facets of the corresponding projected hidden-influence polytopes. These results identify linear-cluster graph states as certifiable and experimentally friendly resources for finite-speed hidden-influence tests.

quant-ph↗

arXiv2026

Learning User-Aware Recall: Personalized Retrieval in Long-Term Conversational Memory

Long-term conversational agents are expected to remember past interactions, but memory is useful only when the right evidence is recalled for the right user. Existing memory-augmented LLM agents have made progress in building compact memory banks, yet retrieval is still often driven by query-centered similarity or fixed ranking rules, leaving user-conditioned relevance underexplored. To address this gap, we propose Profile-guided Personalized Retrieval Optimization (PPRO), a retrieval-centric framework that makes memory retrieval both user-aware and optimizable. PPRO builds episodic and semantic memory banks from dialogue histories and derives a user profile from accumulated memories. The profile serves as an explicit personalized prior in memory ranking, allowing retrieval to account for stable user attributes, preferences, and relationships. PPRO further trains a query rewriter with Group Relative Policy Optimization, using both evidence retrieval quality and downstream answer quality as feedback while keeping the memory banks and answer model fixed. Experiments on LoCoMo and LongMemEval-S show consistent gains over training-free memory systems and training-based baselines. Ablation studies further show that both profile-guided ranking and retrieval-oriented rewriting contribute substantially to performance, highlighting retrieval optimization as a key factor in personalized long-term memory use.

cs.IR↗

arXiv2026

Break the Brake, Not the Wheel: Untargeted Jailbreak via Entropy Maximization

Recent studies show that gradient-based universal image jailbreaks on vision-language models (VLMs) exhibit little or no cross-model transferability, casting doubt on the feasibility of transferable multimodal jailbreaks. We revisit this conclusion under a strictly untargeted threat model without enforcing a fixed prefix or response pattern. Our preliminary experiment reveals that refusal behavior concentrates at high-entropy tokens during autoregressive decoding, and non-refusal tokens already carry substantial probability mass among the top-ranked candidates before attack. Motivated by this finding, we propose Untargeted Jailbreak via Entropy Maximization(UJEM)-KL, a lightweight attack that maximizes entropy at these decision tokens to flip refusal outcomes, while stabilizing the remaining low-entropy positions to preserve output quality. Across three VLMs and two safety benchmarks, UJEM-KL achieves competitive white-box attack success rates and consistently improves transferability, while remaining effective under representative defenses. Our experimental results indicate that the limited transferability primarily stems from overly constrained optimization objectives.

cs.CV↗

arXiv2026

Woodpecker Distillation: Weak Models Diagnose Reasoning Bugs in Strong Models

Large language models often fail on reasoning tasks despite possessing the capability to solve them. We argue that many such failures arise from localized reasoning bugs in intermediate steps rather than from global incompetence. We show that these bugs are frequently repairable: inserting a short patch generated by a weak probe model after the same strong-model reasoning prefix can redirect the trajectory toward a correct solution. However, this corrective effect is not reliably internalized by directly fine-tuning on weak patches or repaired trajectories, suggesting that the useful signal lies not in the intervention text itself, but in how it reshapes the model's future reasoning distribution. We therefore propose Woodpecker Distillation, a weak-to-strong training framework that learns from contrastive local interventions. Our method contrasts successful and unsuccessful weak-model patches at the same prefix, constructs a corrective teacher distribution from their induced future token predictions, and distills this signal into the strong model. Experiments on mathematical reasoning benchmarks show that Woodpecker Distillation consistently improves strong-model performance and outperforms direct imitation baselines.

cs.AI↗

arXiv2026

One Tool Is Enough: Reinforcement Learning for Repository-Level LLM Agents

Locating files and functions requiring modification in large software repositories is challenging due to their scale and structural complexity. Existing LLM-based methods typically treat this as a repository-level retrieval task and rely on multiple auxiliary tools, which often overlook code execution logic and complicate model control. We propose RepoNavigator, an LLM agent equipped with a single execution-aware tool: jumping to the definition of an invoked symbol. This unified design reflects the actual flow of code execution while simplifying tool manipulation. RepoNavigator is trained end-to-end via Reinforcement Learning (RL) directly from a base pretrained model, without relying on closed-source distillation. Experiments demonstrate that RL-trained RepoNavigator achieves state-of-the-art performance, with the 7B model outperforming 14B baselines, the 14B model surpassing 32B competitors, and the 32B model exceeding closed-source models such as GPT-5 on most metrics. These results confirm that integrating a single, structurally grounded tool with RL training provides an efficient and scalable solution for repository-level issue localization.

cs.SE↗

arXiv2026

RepoZero: Can LLMs Generate a Code Repository from Scratch?

Large Language Models (LLMs) have recently shown remarkable progress in code generation, yet their ability to construct complete software repositories from scratch remains poorly understood. A fundamental bottleneck is the lack of verifiable and scalable evaluation: existing benchmarks either focus on patch-based editing or rely on human or LLM-based judgments, which introduce bias and limit reproducibility. In this work, we present RepoZero, the first benchmark that enables fully automated, execution-based verification of repository-level generation from scratch. Our key idea is to reformulate generation as repository reproduction: given only API specifications, an agent must re-implement an entire repository such that its behavior matches the original implementation. This design allows for strict black-box validation via output equivalence, while naturally supporting large-scale construction by reusing existing open-source repositories. To further mitigate data leakage and shortcut solutions, we introduce cross-language constraints and a sandboxed evaluation protocol. Building on this benchmark, we propose an Agentic Code-Test Evolution (ACE) framework that performs iterative test generation and error-driven refinement, enabling effective test-time scaling for repository-level synthesis. Extensive experiments across multiple state-of-the-art LLMs and agent frameworks reveal that even the strongest LLM agents achieve only limited pass rates (30\% - 55\%), exposing a substantial gap between current capabilities and real-world software development requirements. Our results establish RepoZero as a challenging, scalable, and reliable testbed for end-to-end code generation, and highlight self-verification via test generation as a critical direction for advancing LLM-based coding agents.

cs.SE↗

arXiv2026

It Takes 8 Tokens: Weak-to-Strong Off-Policy RL via Auxiliary Branches

Reinforcement learning with verifiable rewards has emerged as a standard approach for enhancing reasoning in large language models, which typically optimizes the policy by contrasting multiple self generated rollouts. However, we identify a critical support limited bottleneck in this paradigm: on challenging reasoning tasks, the target model's samples often exhibit semantic redundancy, converging into the same erroneous "reasoning basins" that offer negligible reward contrast for policy updates. In this paper, we propose to overcome this limitation through a weak to strong learning paradigm, where a policy's exploration is informed by a weaker but computationally efficient auxiliary model. We introduce W2SPO, an off policy RL method that injects short auxiliary segments often as brief as 8 tokens into intermediate target model trajectories and the target model then completes the reasoning path from these diverted states. Policy updates are restricted to these short inserted segments based on final verifiable rewards. Empirically, W2SPO achieves superior performance among evaluated 4B scale models on mathematical reasoning benchmarks, outperforming evaluated post trained baselines. Compared with vanilla GRPO under the same sampling budget, W2SPO improves Pass@1 from 62.3% to 64.2% while achieving a 3.55 times training speedup. These results suggest that weak auxiliary branches can induce stronger target reasoning policies by expanding local exploration support.

cs.AI↗

arXiv2026

Student Guides Teacher: Weak-to-Strong Inference via Spectral Orthogonal Exploration

Large Language Models (LLMs) often suffer from ''Reasoning Collapse'' on challenging mathematical reasoning tasks, where stochastic sampling produces lexical variations of the same erroneous logic rather than genuine semantic exploration. We observe that failed reasoning traces are often associated with a low-rank bias manifold in the model's hidden-state geometry, which reduces exploration toward corrective solution directions. To address this, we propose Spectral Orthogonal Exploration (SOE), a geometric inference framework under a ''Student Guides Teacher'' paradigm. Instead of using a weak auxiliary agent for imitation, SOE uses it as an orthogonal probe to introduce semantically heterogeneous reasoning signals into the teacher's orthogonal complement of its dominant subspace. This intervention steers the teacher toward more diverse reasoning trajectories and improves exploration beyond standard sampling. Experiments on mathematical benchmarks show that SOE improves average accuracy by 62.4\% and average sampling efficiency by 113.7\% over baseline methods, suggesting that geometric interventions can be effective for mitigating reasoning collapse in mathematical reasoning. We further provide preliminary evidence that SOE is also effective on logic and code generation benchmarks.

cs.AI↗