arXiv subjects

Liang Chen

Publications and source records attributed to Liang Chen.

At least 37 records · Page 2Linked to original sources

arXiv2026

J-LAW: Joint Localization and Action-Conditioned World Modeling via Coupled Latent Factor Graphs

Classical simultaneous localization and mapping (SLAM) estimates metric poses and a geometric map but does not provide an action-conditioned predictive state. Action-conditioned world models learn compact latent dynamics but ignore global metric consistency and accumulate drift under open-loop rollout. We introduce J-LAW (Joint Localization and Action-Conditioned World Modeling), a unified factor-graph formulation that connects metric pose variables, predictive latent states, and persistent latent landmarks in this letter.J-LAW represents each image as a compact predictive state and combines it with pose or motion measurements through a separately learned mapping. Its maximum a posteriori (MAP) factor graph enforces consistency between these complementary sources of information over time. Experiments on PushT and WildGS show that J-LAW's factor-graph representation can improve long-horizon latent consistency and recover more reliable predictive states under partial observations, forming a foundation for future integrated localization and planning systems.

cs.RO↗

arXiv2026

CVE-SAI: Counterfactual Visual Evidence-Guided Selective Attribute Indexing for Risk-Controlled E-commerce Search

Multimodal product models can complete missing e-commerce attributes, yet current methods still optimize attribute-answer accuracy without verifying visual support, conflate transient prediction with persistent index admission, and lack explicit risk control over factually incorrect or visually unsupported values. We address these gaps with Counterfactual Visual Evidence-Guided Selective Attribute Indexing (CVE-SAI), which first infers and freezes an ontology-constrained candidate from the primary image and attribute question without catalog text, and then decides whether that candidate should enter the index. Focus-Zone Distortion (FZD) constructs an attribute-specific visual-dependence proxy through a controlled counterfactual intervention, and Evidence-Guided Attention Redistribution (EGAR) uses the proxy to refine ontology-constrained scoring. The canonical candidate is frozen before evidence necessity, evidence retention, nuisance-transformation stability, and candidate-specific catalog-text conflict audits; catalog text can only tighten admission and cannot revise the candidate. Independent family-level calibration selects one policy with a simultaneous one-sided finite-sample bound under a 5% unsafe-admission budget. Experiments on five visual attributes derived from Amazon Berkeley Objects show that CVE-SAI improves attribute inference and evidence localization, achieves the highest certified admission coverage under the shared risk protocol, and yields the strongest controlled retrieval performance with the lowest unsafe auto-induced exposure among automatic-admission systems. Separating inference from admission therefore enables visually supported attribute completion to improve retrieval while limiting persistent index contamination.

cs.AI↗

arXiv2026

One Polluted Page Is Enough: Evaluating Web Content Pollution in LLM Recommenders

Search-augmented LLMs increasingly mediate everyday consumer recommendations by retrieving live web content. This creates a new risk: LLM recommenders may consume web content that Generative Engine Optimization (GEO) operators have polluted to mislead them. We ask: to what extent do they become unwitting promoters of fake products? We introduce FORGE (Fake Online Recommendations in Generative Environments), which locally rewrites real products in a frozen set of retrieved web pages into fake ones and measures how often the LLM recommends the fake product, across 225 real products in 15 categories and 5 consumer scenarios. Across 12 commercial and open-weights LLMs, all models are vulnerable: a single polluted page yields fooled rates of up to 27%, while the full top-3 replacement raises this to 73.8%. Vulnerability varies across categories, increasing when models lack stable prior knowledge of the products. Reasoning does not mitigate this vulnerability; instead, it often generates spurious social proof to justify false recommendations. None of the four defenses is adequate: the skepticism prompt can exacerbate vulnerability much like reasoning, the two consensus filters risk suppressing legitimate products, and credibility re-ranking helps every model but removes only a sixth of the fakes. We release the FORGE benchmark and the evaluation code at https://github.com/leoluolol/forge-benchmark.

cs.CL↗

arXiv2026

DART: Dual-Axis Airborne Reachability-Gated Torque-Reaction for Off-Road Vehicle Jumps

Traversing crests, ledges, and ditches at high speed often launches vehicles into the air, and a mishandled landing presents a substantial crash hazard. We show that the airborne phase is barely controllable: on a 1383 kg platform the wheel angular-momentum budget caps the recoverable pitch-rate change at roughly $9$-$13^\circ$/s in the tighter nose-up direction under drive at typical takeoff wheel speeds, and at about twice that in the reverse-inclusive braking direction; driving the wheels to their drivetrain hard limit raises the measured nose-up ceiling to only $16$-$18^\circ$/s. Takeoff pitch-rate disturbances beyond this directional budget are physically unrecoverable in flight, so the decisive leverage lies before takeoff. DART (Dual-Axis Airborne Reachability-Gated Torque-Reaction) back-propagates the landing constraint into a closed-form certified feasible-takeoff set, which supplies a conservative go/no-go condition and a pre-takeoff speed-shaping law. In flight, DART regulates pitch and roll via steer-resolved wheel-reaction torque, governed by a per-flight roll latch derived from the yaw-coupling analysis. In deterministic full-scale simulation in BeamNG.tech, a calibrated pre-takeoff speed regulator reduces touchdown speed by 36% and raises on-target landings from 0/30 to 30/30. Under the same steep-lip approach the airborne law completes 29/30 safe landings under crash-avoidance bounds versus 0/30 for reaction-wheel-style PD (RW-PD) and time-optimal bang-bang (TOBB). On banked run-ups DART holds the median pitch error at or below $2^\circ$ at every cross-slope, with the largest baseline separation at $γ=12^\circ$. Across disturbance regimes, the latch preserves pitch-only allocation on low-disturbance entries and enables dual-axis control when roll becomes binding. All results are from simulation; hardware validation remains open.

cs.RO↗

arXiv2026

DART-S: Reachability-Audited Active-Suspension Preconditioning for Off-Road Vehicle Jumps

Airborne torque reaction cannot recover takeoff errors beyond the wheel angular-momentum budget. DART-S applies ramp-face suspension preconditioning to change pitch, pitch rate, and wheel spin before liftoff, thereby shifting the queried state and altering the remaining authority budget. To predict how each suspension action reshapes this state-budget pair, DART-S employs a local calibration map. A support-aware selector combines the predicted shift with local outcome evidence and an interval-reachability screen; an exact-pair audit reports residual authority. Across 600 new runs in 72 independent BeamNG sessions, every positive, negative, and boundary query follows its prespecified branch. At the confirmed 40°/13 m/s boundary, DART-S attains 24/24 post-touchdown attitude-criterion successes versus 0/24 for DART (session-level Holm-adjusted p=0.0234). At 11.5 m/s, a 0.35 s timing action attains 23/24 versus 0/24 for the static preset (p=0.0156). The 200 rad/s command guard keeps drivetrain hard-limit exceedance at zero across all 600 runs. The source code will be available at https://github.com/MeridianCAS/DART-S

cs.RO↗

arXiv2026

Hundred-hertz quantum circuit iteration rate in a reusable neutral-atom array

Neutral-atom quantum processors have rapidly advanced in scale and coherence, yet their practical performance remains constrained by limited quantum circuit iteration rates (qCIRs) and information throughput. Here we experimentally demonstrate a high-throughput neutral-atom system based on non-destructive readout and atom reuse. By integrating a chip-based photonic interface with a 10-qubit array, we implement non-destructive readout with a retention probability of 99.7%, and further achieve a raw qCIR of 101Hz and a post-selected qCIR of 74.8Hz. More importantly, we verify a general throughput optimization methodology and obtain a normalized Fisher information rate of 57.7Hz, improving the achievable throughput by more than one order of magnitude compared with conventional methods. Our results establish a practical route toward high-throughput neutral-atom quantum processors.

quant-ph↗

arXiv2026

Magnetic rigidity reveals the PeVatron acceleration region in SS 433

PeVatrons are cosmic accelerators capable of driving particles to petaelectronvolt (PeV) energies. Recently, microquasar jets have emerged as compelling Galactic PeVatron candidates. This is especially the case for SS 433 as its $>100$ TeV gamma-ray emission is spatially coincident with an atomic cloud. However, the exact region where PeV protons are accelerated and injected within these jets remains unresolved. Here we report, using archival, multi-frequency VLBA observations, the magnetic field profile $B(H)$ along the SS 433 inner jet on tens of AU scale, where $H$ is the distance from the central compact object. We find that the field declines as $B(H) \propto H^{-0.50\pm0.12}$, demonstrating that the magnetic rigidity $B(H)R_{\rm acc}$ grows with $H$ for a conical jet. This implies the Hillas limit ($E_{\rm max} \propto BH$) to lie well beyond a PeV at a few hundred-AU scale, which becomes a highly potential site for accelerating protons to energies $E_{\rm cut} \simeq 2.6$ PeV inferred from the LHAASO gamma-ray spectrum. These results reveal a hidden PeVatron within the baryonic ejecta of microquasar SS 433, well upstream of the extended TeV-emitting lobes.

astro-ph.HE↗

arXiv2026

Jet Power, Bulk Lorentz Factor, Black Hole Spin, and Magnetic Field of Accretion Disk in Jetted Active Galactic Nuclei: A Large Gamma-Ray Emission Sample

We present a catalog of physical parameters for powerful jet-accretion disk-black hole systems in one of the largest samples of gamma-ray emitting jetted active galactic nuclei (AGNs), including jet kinetic and radiative powers, jet radiative efficiencies, bulk Lorentz factors, black hole spins, accretion-disk magnetic fields and Compton dominance. Comparing jet kinetic power estimators for blazars, values derived from spectral energy distribution (SED) fitting tend to exceed those estimated via cavity power and other scaling relations. For radiatively efficient AGNs, most sources are inferred to possess high spins; for radiatively inefficient AGNs, many potentially have high spins, though some may differ. This indicates that black hole spin does not effectively distinguish radiatively efficient from inefficient jetted AGNs. Our results suggest accretion-disk magnetic field strength as a key discriminator, proposing a tentative dividing value of $\approx 10^{3.9}$ Gauss between radiatively efficient and inefficient populations. Jet power and bulk Lorentz factor exhibit significant correlations with black hole mass in radiatively efficient AGNs, while weak-to-moderate correlations are observed in radiatively inefficient AGNs within narrow accretion-rate bins. Our analysis reveals that jet power correlates with both disk luminosity and magnetic field strength. Furthermore, correlations linking Eddington ratio and Compton dominance with jet properties are consistent with the jet-accretion connection. Finally, jet radiative power and bulk Lorentz factor show a potential dependence on black hole spin. These results are consistent with the scenario in which jets are powered and accelerated by energy extraction from rapidly spinning black holes via accretion-disk magnetic fields.

astro-ph.HE↗

arXiv2026

A scalable chip-integrated single-photon source array based on 50 individually addressable neutral atoms

Scalable arrays of identical single-photon sources are a central resource for photonic quantum information processing, quantum networks and quantum metrology. Neutral atoms provide intrinsically identical emitters that can be assembled and rearranged in optical tweezers, but a many-channel fiber interface to individually trapped atoms has remained a major technical challenge. Here we demonstrate a chip-interfaced single-photon source array based on 50 individually addressable $^{87}\mathrm{Rb}$ atoms. A glass waveguide fan-out converts the \SI{5}{\micro m} pitch of the optical-tweezer array to the \SI{127}{\micro m} pitch of a commercial fiber array, mapping each atom to its own waveguide, fiber and single-photon detector. We resolve all 50 channels with an average nearest-neighbor cross-talk of $0.4\%$ and a uniform insertion loss of \SI{2.9}{dB}, and verify single-photon emission with $g^{(2)}(0)=0.29$, presently limited by detector dark counts and residual cooling-light scattering. Combining per-channel atom discrimination, rearrangement and reservoir replenishment, we prepare source subarrays of up to 24 atoms with a $93\%$ fill fraction. For small target numbers, atom loss is repaired from the reservoir at the detection-limited rate of \SI{118}{Hz}. We further fabricate a 784-channel waveguide chip, showing that the photonic interface can be extended well beyond the present number. This architecture establishes a fiber-native neutral-atom platform for larger arrays of identical single-photon sources.

physics.atom-ph↗

arXiv2026

MedClaw: Heuristic Agent Harness for Long-Horizon Surgical Video Reasoning

Understanding tens-of-minutes surgical videos requires long-horizon temporal reasoning, answering what happens before, after, or across stages of a procedure by grounding the question in visual evidence spread across time. Existing approaches handle this poorly: a one-shot vision-language model (VLM) compresses the whole procedure to fit its context window and loses the detail a "before" or "after" question depends on, while video agents that train the model where to look are data-hungry and transfer poorly to out-of-domain surgery. We build an agent harness that separates reasoning from perception and improves by evolving context rather than optimizing weights. A text-only orchestrator plans which evidence to gather and issues an auditable sequence of tool calls, while frozen vision-language sub-agents execute each call over the pixels, viewing, cropping, inspecting frames, and retrieving external knowledge. We further propose a gradient-free, reward-gated Heuristic Skill Distillation loop that mines the agent's own low-scoring traces and keeps a candidate skill only when it raises a validation reward, yielding reusable retrieval skills, notably directed re-look. Growing an external skill library rather than tuning weights, the loop adapts from only about 100 labeled examples, far fewer than supervised or reinforcement fine-tuning requires. To evaluate this agent, we introduce MedClawBench, a de-leaked, doctor-grounded benchmark of 1,123 questions over self-built long neurosurgery recordings and a held-out public lecture-video test split. Across both datasets and all four evaluation dimensions, our agent consistently outperforms one-shot VLMs and general video-agent frameworks, with the largest gains on the long, out-of-domain neurosurgery videos. Project page: https://fyycs.github.io/medclaw/.

cs.CV↗

arXiv2026

MindMemOS: A Portable and Self-Evolving Memory Operating Layer for AI Agents

Memory is a core component of AI agents, enabling them to accumulate experience, maintain personalization, and adapt over long-term interactions. However, existing memory systems often remain fixed after development, limiting their ability to adapt their memory models, organization strategies, and procedural knowledge through continued use. We present MindMemOS, a portable and self-evolving memory operating layer that organizes open-world information using a unified entity property timestructure. MindMemOS supports scenario-adaptive memory modeling, higher-order pattern discovery, autonomous memory refinement, and continuous skill evolution. Its MindMemEvolve algorithm employs validation-driven evolutionary search to optimize memory schemas for target scenarios, whiledreaming consolidates accumulated memories by merging redundant records and resolving conflicts. In addition, implicit corrective feedback serves as a human-in-the-loop signal for identifying and revising potentially inaccurate or misaligned memories. Its MindSkillEvolve algorithm further transforms agent execution trajectories into reusable and progressively refined skills. MindMemOS achieves 94.03% accuracy on LOCOMO and 70.63% on PersonaMem. MindSkillEvolve improves SpreadsheetBench success by 9.2 percentage points over the initial-skill baseline.

cs.AI↗

arXiv2026

Anisotropic Particle Transport from a Pulsar Wind Nebula Revealed by Einstein Probe and LHAASO

Pulsar wind nebulae (PWNe) are major cosmic ray accelerators, yet the mechanisms transporting high-energy particles into the interstellar medium remain elusive. Building on the LHAASO discovery of an ultra-high-energy (UHE) $γ$-ray source near the bow-shock PWN powered by the pulsar PSR J1740+1000, we present a joint Einstein Probe (EP) and LHAASO study of this system. EP observations reveal an extended X-ray tail far exceeding the structure previously seen by XMM-Newton. Updated LHAASO observations show that the $γ$-ray emission is elongated, with its major axis aligned with the extended X-ray tail revealed by EP. This is the first detection of an X-ray pulsar tail associated with a spatially coincident extended UHE $γ$-ray emission. The X-ray and $γ$-ray spectrum can be well explained with a single population of relativistic electrons via synchrotron and inverse Compton radiation, respectively, removing the need for particle re-acceleration during propagation. The results unambiguously show that electrons/positrons above 100 TeV are escaping from the PWN. Instead of the immediate, isotropic diffusion into ambient interstellar medium that is typically assumed, these particles are transported anisotropically over at least $\sim$10 pc, either guided by the background magnetic field or carried by an advective outflow.

astro-ph.HE↗

arXiv2026

Kimi K2.5: Visual Agentic Intelligence

We introduce Kimi K2.5, an open-source multimodal agentic model designed to advance general agentic intelligence. K2.5 emphasizes the joint optimization of text and vision so that two modalities enhance each other. This includes a series of techniques such as joint text-vision pre-training, zero-vision SFT, and joint text-vision reinforcement learning. Building on this multimodal foundation, K2.5 introduces Agent Swarm, a self-directed parallel agent orchestration framework that dynamically decomposes complex tasks into heterogeneous sub-problems and executes them concurrently. Extensive evaluations show that Kimi K2.5 achieves state-of-the-art results across various domains including coding, vision, reasoning, and agentic tasks. Agent Swarm also reduces latency by up to $4.5\times$ over single-agent baselines. We release the post-trained Kimi K2.5 model checkpoint to facilitate future research and real-world applications of agentic intelligence.

cs.CL↗

arXiv2026

Controlled Memory Interference in Continual LLM Agents

Long-term memory enables AI agents to maintain continuity across sessions, personalize behavior, and evolve through accumulated experience. Yet memory evolution is not simply a process of storing more information: new experiences may reinforce, revise, or interfere with existing memory states. Existing systems mainly emphasize memory construction and relevance-based retrieval, but several memories may remain simultaneously relevant while differing in state, temporal validity, or authority. We introduce Controlled Memory Interference (CMI), a controlled diagnostic and data-generation framework for studying how agent memory evolves under different memory relationships. Across controlled memory evolution, benign accumulation has limited effects, whereas relationship-specific interference sharply suppresses update plasticity with little stability gain, either by blocking target-memory exposure or by disrupting its downstream use. Lexical and Dense retrieval exhibit distinct interference pathways, while poisoning is more sensitive to update-authority cues than to recency alone. Beyond diagnosis, CMI provides targeted examples for interference-aware memory learning, improving the distinction between valid updates and interference-inducing memories while preserving performance on original memory tasks. These findings show that memory evolution is shaped not only by memory scale, but also by interactions among accumulated experiences. More broadly, memory interference emerges as an important factor for reliable continual agent memory systems.

cs.AI↗

arXiv2026

Verifiable Memory: Learning Unified Memory Management with Local and Global Verifiers for Large Language Model Agents

Large language model (LLM) agents must retain reusable information, control a bounded active context, and recover earlier evidence during long-horizon interaction. Existing methods commonly optimize long-term memory (LTM) and short-term memory (STM) separately, while unified policies are often trained primarily with trajectory-level feedback, which provides weak credit for individual memory decisions. We present Verifiable Memory (VerMem), a framework that represents LTM, active context, and episodic history as distinct states and controls them with one memory operation policy. Seven atomic operations let the policy add, revise, or soft-delete LTM entries; retrieve LTM into the active context; filter or summarize the active context; and restore selected episodic fragments. VerMem is initialized by supervised fine-tuning and trained with a three-stage reinforcement-learning curriculum. The local verifier scores executable memory transitions, and a global verifier assesses evidence coherence and terminal-memory consistency after task completion. These scores are combined with programmatically computed task, evidence-recall, efficiency, and constraint signals through hierarchical credit assignment. The verifiers are used only during training. Across five benchmarks and two LLM backbones, VerMem achieves the best result on the vast majority of reported metrics and consistently outperforms strong memory baselines. Under controlled online-token budgets on three interactive benchmarks, it also achieves the strongest efficiency--performance frontier among the compared methods. Code is available at https://github.com/Sun-SYSU-24/VerMem.

cs.AI↗

arXiv2026

The Extended Ultrahigh-energy Gamma-Ray Emission in the Vicinity of PSR J2238+5903

We present a comprehensive analysis of the recently discovered TeV gamma-ray source, LHAASO J2238+5900. Based on data collected from the LHAASO, our fitting results suggest that the source is significantly extended with an angular extension of 0.54° \pm 0.01° and is spatially coincident with the pulsar PSR J2238+5903. Its spectrum is characterized by a power-law with a cutoff at 41.0\pm 3.5 TeV. Additionally, the source exhibits a significant signal of 7.9σabove 100 TeV, implying that it is a PeVatron candidate. While the gamma-ray emission is consistent with a pulsar wind nebula (PWN) scenario, the relatively large extension size also allows for a halo interpretation, potentially caused by electron-positron pairs escaping from the PWN.

astro-ph.HE↗

arXiv2026

Rydberg quantum antennas for chip-interfaced single-photon source array

We propose a Rydberg quantum antenna, consisting of a one-dimensional chain of neutral atoms trapped in a standing-wave optical tweezer, as a chip-interfaced single-photon source. Rydberg blockade induces a single collective excitation in the atomic chain, while its ordered geometry shapes the photon emission into a directional beam, thereby realizing a quantum antenna that emits strictly one photon at a time. Through numerical simulations, we demonstrate single-photon collection efficiencies into a waveguide exceeding 70\% through a high-NA lens with a negligible multiple excitation probability. These performance are robust against probabilistic atom loading in the standing-wave dipole traps, requiring only ten atoms, indicating experimental feasibility. The Rydberg quantum antenna thus offers a unique route toward scalable arrays of high-efficiency, high-purity, and identical single-photon sources interfaced with a photonic chip, addressing key challenges in photonic quantum information technology.

quant-ph↗

arXiv2026

ClawBench: Can AI Agents Complete Everyday Online Tasks?

AI agents may be able to assist with emails and documents, but can they reliably complete everyday online workflows on real websites? Everyday online tasks offer a realistic yet unsolved testbed for evaluating the next generation of AI agents. To this end, we introduce ClawBench, an evaluation framework comprising 153 everyday online tasks that people need to accomplish regularly in their lives and work, spanning 144 platforms across 15 categories, from completing purchases and booking appointments to submitting job applications. These tasks require capabilities beyond existing benchmarks, such as obtaining relevant information from user-provided documents, navigating multi-step workflows across diverse platforms, and write-heavy operations like filling in many detailed forms correctly. Unlike existing benchmarks that evaluate agents in offline sandboxes with static pages, ClawBench operates on production websites, preserving the full complexity, dynamic nature, and interaction challenges of real-world web environments. An interception layer captures and blocks the final submission request, ensuring safe evaluation without real-world side effects. Our evaluations of 8 frontier models show that both proprietary and open-source models complete only a small portion of these tasks. For example, Claude Sonnet 4.6 achieves only 33.3%, which exposes gaps in current AI agents. Progress on ClawBench brings us closer to AI agents that can function as general-purpose assistants.

cs.CL↗