Searcharxiv⌕ Search

arXiv subjects

Yong Yu

Publications and source records attributed to Yong Yu.

At least 19 recordsLinked to original sources
arXiv2026

A pilot study on the CSST astrometric capability: Detecting astrometric binaries with Gaia synergy via simulated data

Context. The China Space-station Survey Telescope (CSST) will provide deep, wide-field epoch astrometry during its 10-year mission. Astrometric binary orbits constrain the masses of stellar and compact-object components. Orbital recovery depends on astrometric precision and temporal coverage. Combining CSST and Gaia data extends the baseline and improves binary detection. Aims. We evaluate CSST, Gaia, and joint astrometry for binary-candidate selection and 12-parameter (12p) orbit fitting at faint magnitudes ($g>17.8$). We also test how regular CSST cadences affect the yield of 12p fits satisfying our criteria. Methods. We constructed a mock catalog, simulated CSST and Gaia epoch astrometry, and fitted five-parameter (5p) single-star models to derive astrometric diagnostics, proper-motion anomaly features, and observational-sampling features. A four-stage histogram-based gradient-boosting classifier used these features to select candidates for 12p orbit fitting and assessment. Results. On the independent test set, the classifier reaches a precision of 0.802 and a recall of 0.181 among eligible true binaries. In the scenario-specific fitted samples, joint astrometry raises the fiducial fraction from 6.76% for Gaia alone to 10.46%; for fitted binaries with $P_{\rm true}>15{\rm yr}$, it rises from 2.37% to 6.78%. The current CSST schedule yields few fiducial fits, while idealized regular cadences increase the yield mainly at $g\lesssim21$. Conclusions. In the simulation, joint CSST and Gaia epoch astrometry yields higher fractions of fitted unresolved binaries satisfying the stated criteria than Gaia-only solution. A practical strategy is to select candidates from 5p diagnostics and astrometric anomalies, obtain more regular CSST follow-up observations, and then fit 12p orbital models and apply the selection criteria.

astro-ph.IM↗
arXiv2026

Efficient Conversion of Optical to Mechanical States Close to the Single-Quantum Level

Coherent interfaces between optical photons and mechanical excitations provide a promising route towards phonon-state engineering and hybrid quantum information processing. Cavity optomechanical systems enable such interfaces via optomechanically induced transparency (OMIT), allowing coherent mapping between traveling optical fields and localized mechanical modes. However, previous implementations of OMIT conversion protocols were limited to classical input signals with large coherent state photon numbers due to room temperature operation and corresponding thermal mechanical noise. Here, we demonstrate efficient low-noise photon-phonon state transfer close to the single-quantum regime in an optomechanical crystal operated at Millikelvin temperatures. Using weak coherent optical input pulses at the few-photon level, we achieve a record-level photon-phonon conversion efficiency of $\eta=0.76$, a mechanical storage lifetime of $T_\mathrm{1}=7.3~\mu s$, and a tunable conversion bandwidth exceeding 4.5 MHz. Hanbury Brown-Twiss measurements of the retrieved signal demonstrate the coherent nature of the converted phononic state, evidencing low added thermal noise in the conversion process ($n_\mathrm{th}=9.0$). These results establish optomechanical crystals as efficient optical interfaces to GHz mechanical modes and provide a pathway toward deterministic single-quantum-level mechanical state preparation.

quant-ph↗
arXiv2026

From Final Artifacts to Trajectories: Retrospective Process Supervision for Evidence-Grounded Long-Form Generation

Trajectory data is getting more vital for training large language models for boosting the agentic abilities. Unlike the verifiable domains such as coding or mathematics, scaling trajectory data for open-ended tasks is much more difficult because these tasks lack singular ground truth and are costly to annotate or verify. In this paper, we propose RetroGen, a self-improving framework of retrospective process supervision. Our key observation is that although expert trajectories are scarce, high-quality final artifacts such as literature reviews, analyst reports and legal judgments, are abundant in pre-training data and can be viewed as compressed traces of the evidence-seeking processes that produced them. RetroGen reconstructs candidate latent trajectories from expert artifacts, verifies them against both the artifact and supporting evidence, and trains models on their own successful reconstruction data, without requiring trajectory data from stronger models. Experiments show that RetroGen improves grounding, faithful synthesis, and long-form evidence-seeking agent tasks.

cs.CL↗
arXiv2026

What Makes Good Agentic Data? An ACE Lens on Data Generation for LLM Agents

LLM agents increasingly rely on generated interaction data to learn how to interact with external environments. Agentic data generation must maintain consistency among environments, tasks, interactions, and success signals while producing experience that is useful rather than merely abundant. Existing work spans many agent domains, but domain-centered organization and heterogeneous evaluation often obscure common generation mechanisms and conflate candidate construction with verification and selection. This work develops a two-level framework for the field. First, we represent agentic data as a common factorized object $(E,q,\tau,v)$, comprising an environment specification, task signal, interaction realization, and optional verifier. We organize generation paradigms by their primary anchor and dependency structure. Second, we formulate generation as constrained distribution design through the Accuracy-Complexity-divErsity (ACE) lens. Accuracy establishes the feasible support of grounded and internally consistent data. Within this support, Complexity places learning mass relative to the capability of a declared learner and execution configuration, while divErsity controls coverage and redundancy of data. Using this framework, we explore how prior work verifies generated experience, constructs and calibrates difficulty, and expands behavioral coverage. The literature reveals a shift toward execution-grounded accuracy, learner-relative complexity, and diversity beyond surface variation or dataset size. We further discuss broader directions and emerging trends in agentic data generation through the ACE lens, including their implications for scaling, data sources, training regimes and adaptive learning. Overall, the central challenge is not simply to generate more data, but to continually allocate valid, informative, and non-redundant experience as agents and environments evolve.

cs.AI↗
arXiv2026

SEAM: Shot Entity-Attribute Memory for Consistent Short-Drama Generation at Scale

Short-drama generation has grown into a large, industrialized pipeline, and as it scales from isolated shots to the episode level, visual continuity has become a critical bottleneck. Current agent frameworks generate each shot in isolation, so context drifts across shots and props, character posture, and blocking turn inconsistent. Once assembled, these small discrepancies amplify into severe visual breaks. We present SEAM (Shot Entity-Attribute Memory), a training-free, model-agnostic memory graph that repairs continuity entirely at the prompt-text layer by extracting a multi-dimensional state for every shot, retrieving only causally prior context over the resulting graph, filtering it selectively, and injecting the surviving constraints by natural-language prompt rewriting. We further release SEAM-Bench, a double-blind continuity storyboarding benchmark, on which SEAM raises cross-episode continuity recall from 0.700 to 0.946, generalizes across six mainstream text models, and yields consistent, though not yet significant, gains at the generated-image layer. Deployed as a mandatory stage in CreativeFitting's SEAM-Agent production pipeline over 201 shots, SEAM reaches a 96.5% director-acceptance rate with zero unsafe injections; a conservative counterfactual attributes at least 21.9 percentage points of that rate to its cross-episode memory.

cs.AI↗
arXiv2026

SkillGate: Training In-Policy Skill Selection in Long-Horizon Agents

Agent frameworks increasingly package procedural knowledge as skills: instruction files an agent reads on demand, while public libraries now hold thousands of them. Which skill to read has thus become a decision the policy itself makes in the middle of an episode, yet no existing signal trains it. We show that the default remedy, outcome-rewarded RL over the candidate slate, cannot teach it, for a structural reason we identify and name selector credit starvation: under a broadcast, sequence-level advantage, the few tokens that name the chosen skill carry a vanishing share of the loss, and the credit they inherit is increasingly wrong-signed as trajectories lengthen. A correct choice is punished whenever the execution after it fails, even though the choice itself is among the most valuable decisions in the trajectory. Auditing a completed run's own training artifacts confirms all three properties, each worsening monotonically with horizon. SkillGate removes the failure by construction: it partitions the token support into two disjoint credit channels, outcome credit reaching only execution tokens, and a separate action-local advantage reaching exactly the skill-naming tokens, positive only when a trajectory's single read is the correct one. On five agentic benchmarks under a 16-candidate slate, SkillGate lifts a 9B policy from 40.8% to 53.2% trial success, well ahead of the identical budget spent on outcome reward alone, while cutting exposure to misleading candidates by two thirds and reading fewer skills.

cs.AI↗
arXiv2026

SAGE: Self-Evolving Storyboard Skills via Attribution-Guided Rule Evolution

Storyboards turn screenplays into visual shot plans for automated short drama production. Professional storyboarding relies on tacit directorial expertise and remains an industrial bottleneck. Large language models can automate this step, but methods for supplying directing knowledge face three challenges: (1) Knowledge acquisition: the craft remains implicit in exemplars or must be written manually. (2) Knowledge refinement: authored knowledge is not evaluated against execution outcomes, and opaque generation prevents feedback attribution to the knowledge behind each decision. (3) Knowledge injection: injecting all knowledge exceeds usable context, while manual selection for every narrative group does not scale. We present SAGE (Skill with Attribution-Guided Evolution), a deployed framework that learns, attributes, evolves, and routes directing knowledge from expert demonstrations. SAGE derives rules that are independent of episode content by contrasting each training screenplay with its expert storyboard. During generation, the model records each narrative group's adopted rules. Combining these records with localized feedback enables targeted updates to individual rules. Evolved rules form scenario packages with a routing index, so each group retrieves only a bounded set appropriate to its situation without expert intervention. On 18 test episodes across three genres, SAGE scored 77.8 on a rubric validated by experts, versus 77.1 for professional directors. Deployed for 14 days on Virtual Film Studio, SAGE produced 1,344 narrative group outputs; 87.2 percent were accepted without substantive edits, and the production team recorded over 83 percent less authoring time per episode. We release PROSE, the first public dataset pairing screenplays with storyboards by professional directors across 68 episodes: https://github.com/creDreams/PROSE.

cs.AI↗
arXiv2026

Learning from Unreachable Rewards: Hint-Conditioned Reinforcement Learning for Generative Recommendation

Semantic-ID generative recommenders represent each item as a short sequence of discrete semantic tokens and predict the next item by autoregressively generating this token sequence. This paradigm enables a unified generation interface for item IDs, histories, and item text, but it also creates a structured optimization bottleneck during reward-based post-training: when an early semantic token enters the wrong branch of the item-token space, finite rollout groups rarely reach the ground-truth item, so group-relative optimization receives identical zero rewards and produces no useful advantage. We propose Hint-Conditioned Generative Recommendation (HCGRec), a semantic-ID generative recommendation framework that recovers learning signal for such hard training instances. HCGRec diagnoses each instance with checkpoint rollouts and supplies a minimal target-prefix hint only when the current generator cannot reach the correct item. The model then generates the unhinted suffix under the hinted semantic branch, turning zero-reward groups into informative comparisons over item-token completions. Hinting also changes token identity: hinted prefix tokens are oracle-provided item context, while unhinted suffix tokens are sampled generation actions. We therefore introduce hint-aware credit decomposition, using supervised learning to preserve item-semantic and prefix-structure alignment for hinted tokens and GRPO to optimize the sampled suffix. Experiments on sequential recommendation benchmarks show that HCGRec substantially improves over supervised fine-tuning and vanilla reward-based post-training, while reducing zero-advantage training samples from over 70% to below 20%. The code is accessible at https://github.com/WncFht/GRec.

cs.IR↗
arXiv2026

MARA: Flow-Matching-Guided Multi-Agent Resource Allocation for Computational Resource Efficient Learning

Allocating limited computation among concurrent learning tasks is difficult when each task must reach a target loss before a deadline but its required training effort is unknown. Existing approaches combine online loss prediction with adaptive resource allocation, yet commonly treat computation as continuously divisible throughput. We instead study a practical setting in which tasks arrive over time and computation is provided by discrete nodes. This setting introduces both uncertain demand and constrained sequential decisions. We propose MARA, which predicts future loss trajectories with conditional flow matching and coordinates compute nodes through a cooperative multi-agent autoregressive policy. A potential-based progress reward supplies intermediate training feedback while preserving the undiscounted task-completion objective. Across in-distribution, reinforcement-learning, and vision workloads, flow matching reduces remaining-resource prediction error relative to weighted least squares. At the scheduler's training load, MARA completes 63.46% of tasks on average, 8.54 percentage points above strong baseline Learning with Adaptive Resource Allocation (LARA), and remains ahead under unseen heavier workloads.

cs.LG↗
arXiv2026

Rethink Before You Execute: Adaptive Execution for World Action Models

World Action Models (WAMs) jointly predict future actions and the evolution of the environment. At each inference, a WAM generates a chunk of actions and the robot executes a fixed prefix before replanning. We argue that this fixed execution horizon is poorly matched to execution dynamics: the chunk reliability varies across task stages, so when to replan depends on the result of accumulated execution, not on the step counts. We propose TempoWAM (Timing Execution by Monitoring Progress Online), a lightweight plug-and-play execution scheme for WAMs. A Recurrent Progress Monitor first estimates task progress from the current observation, task instruction, remaining actions, and execution history; and an Adaptive Execution Protocol then evaluates whether the chunk is advancing the task to decide if replanning is needed. To bridge the training-deployment gap, the protocol is calibrated by a task-dependent calibration factor with online adaptation. Experiments on LIBERO, RoboTwin, and real-world tasks show that TempoWAM consistently improves the efficiency-success trade-off of WAM execution. On real robots, it reduces WAM inferences by 26.9% on easy tasks while maintaining success, and improves success by 13.3 points on difficult tasks.

cs.RO↗
arXiv2026

Towards optimal photometric calibration of digital astronomical plates with deep learning

Photometric calibration of digitized photographic plates is commonly modeled with separable magnitude-, color-, and position-dependent terms, but this separability can break down when image quality varies across the field in a magnitude-dependent way, leaving coupled spatial systematics in the residuals. We introduce a deep-learning calibration framework, the Multi-Feature Fused Network (MFF-Net), which takes instrumental magnitude, color, and pixel coordinates as input and learns a single nonlinear correction that jointly captures their coupled dependencies. Tests on 1{,}200 digitized Chinese plates show that MFF-Net consistently outperforms the MYX25 method (Ma et al. 2025), improving the 5th--95th percentile precision from 0.11--0.26~mag to 0.08--0.18~mag and delivering an approximately factor-of-two gain for bright sources. The learned correction largely removes the magnitude--position coupling seen in post-calibration residual maps, enabling higher-precision plate photometry and more reliable use of large historical plate archives.

astro-ph.IM↗
arXiv2026

VAD: Attributing Visual Evidence for Target Reconstruction in Multimodal On-Policy Distillation

Multimodal on-policy distillation (OPD) transfers fine-grained visual knowledge by supervising student-generated trajectories with a privileged-view teacher. Yet its next-token corrections are source-mixed, combining visual signals with linguistic priors and teacher-specific effects. The key challenge is to estimate which corrections are supported by visual evidence, not merely where or how strongly to distill. We introduce Visual Attribution Distillation (VAD), a counterfactual target-reconstruction algorithm that estimates the visually attributable part of a teacher correction. At each student-generated prefix, VAD evaluates the same fixed teacher with the relevant evidence present and removed. The corresponding change in centered log-probabilities defines ut, a signed proxy for the visual evidence direction that estimates how revealing the evidence supports or refutes candidate tokens. VAD projects the original correction onto this proxy to obtain an intervention-aligned component and a proxy-unexplained residual, then reconstructs a student-anchored target from the former. During training, this reconstructed target supplies the primary supervision signal, while the privileged teacher contributes a weak regularizer. Across six fine-grained visual benchmarks at 4B and 9B scales, VAD outperforms direct privileged-view distillation and visual-advantage weighting. Token- level and controlled-target analyses show that the proxy-aligned component is enriched in task-relevant visual corrections and yields stronger target shifts, especially when evidence refutes a mistaken answer. These results support counterfactual target reconstruction as an effective alternative to source-mixed supervision.

cs.CV↗
arXiv2026

The planar Hopf algebra of noncommutative multi-indices

We construct the planar Linares--Otto--Tempelmayr Hopf algebra, thereby filling the missing planar noncommutative multi-index corner in the square relating the LOT, Butcher--Connes--Kreimer, and Munthe-Kaas--Wright Hopf algebras. Starting from the free associative algebra on a weighted alphabet $\mathbb Z_{\ge -1}\times A$, we define an insertion-type product yielding a post-Lie structure on the Lie algebra generated by the linear span $V(A)$ of weight $-1$ monomials whose proper left prefixes all have nonnegative weight, and the Guin--Oudom construction then produces the planar LOT Hopf algebra. We introduce a planar tree fertility map from decorated planar rooted trees to monomials in $V(A)$, prove that it is a linear isomorphism, and obtain a natural Hopf algebra isomorphism with the Munthe-Kaas--Wright Hopf algebra. We further derive an explicit coproduct formula in terms of left-admissible cuts, establish the extraction-contraction coproduct, and construct a word symmetrization operator compatible with the classical tree symmetrization operator.

math.CO↗
arXiv2026

A unified framework for determining transition dipole polarization in solid-state spin defects

Spin-photon interfaces based on solid-state defects are key building blocks for scalable quantum networks and hybrid quantum platforms. Optimizing light-matter coupling in these systems requires precise knowledge of the optical transition dipole polarization, yet for many promising quantum emitters this quantity is hard to determine and therefore remains poorly characterized. Here, we develop a framework for reconstructing electric transition dipole polarization in spin-1/2 solid-state defects directly from ensemble spectroscopy. The approach combines the response of photoluminescence spectra to magnetic field, optical polarization, and strain. Applied to erbium ions in silicon, a particularly challenging system containing multiple crystallographic subsites, the framework identifies strain-induced shifts as the origin of asymmetric ensemble spectra and enables simultaneous determination of the optical dipole polarization and strain-orbital coupling tensor. The resulting model predicts how cavity-ion coupling depends on crystallographic orientation and magnetic-field direction, which we verify using single erbium ions coupled to a nanophotonic cavity. Together, these results establish a broadly applicable route for extracting microscopic properties of solid-state quantum emitters from ensemble spectroscopy and for engineering optimized spin-photon and spin-phonon interfaces.

quant-ph↗
arXiv2026

BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation

Hallucinations remain a major obstacle to deploying large language models (LLMs) in knowledge-intensive settings, where generated responses must be faithfully grounded in provided evidence. Reinforcement learning (RL) is a promising direction for hallucination mitigation, but response-level faithfulness rewards suffer from a granularity mismatch: localized hallucinations can cause supported content to receive spurious penalties. Although recent work introduces fine-grained feedback such as claim-level verification and token-level rewards, unbalanced credit assignment can still induce length, verbosity, or optimization-noise biases. We propose BALTO, a Balanced Token-level Policy Optimization framework for hallucination mitigation. BALTO extracts checkable factual claims, verifies them against the reference context, and projects claim-level judgments to token-level labels. A balanced token-level credit assignment mechanism is introduced into the framework. This design redistributes probability mass from unsupported content toward faithful content, rather than suppressing the entire response. We systematically analyze the limitations of response-level rewards from a theoretical standpoint, and prove BALTO's advantages in training stability and optimization efficiency for hallucination mitigation. Experiments on ConFiQA, RAGTruth, and FinLLM-Eval show that BALTO achieves the highest faithfulness across all six model--benchmark settings and consistently outperforms existing post-training baselines in Q-Score, demonstrating a stronger faithfulness--informativeness trade-off.

cs.CL↗
arXiv2026

DiffCold: A Diffusion-based Generative Model for Cold-Start Item Recommendation

Cold-start item recommendation remains a persistent challenge in real-world systems due to the absence of interaction histories. While prior models attempt to bridge this gap using item content features, they universally suffer from the \textbf{seesaw dilemma}: enhancing performance for cold items inevitably degrades performance for warm items, and vice versa. We identify that this dilemma stems from a fundamental \textbf{distributional disparity}: warm item embeddings occupy a complex ``behavioral manifold" shaped by rich interaction signals, whereas cold item embeddings are constrained to a ``semantic manifold" derived solely from auxiliary content. Existing methods often force a rigid mapping between these inconsistent spaces, causing the model to sacrifice the precision of warm representations to accommodate cold ones. To address this, we propose \textbf{DiffCold}, a diffusion-based generative model that unifies warm and cold representations. Unlike GANs or VAEs, DiffCold leverages conditional diffusion to reconstruct warm item embeddings from content, preserving the underlying manifold structure without degradation. We further tailor this paradigm with two specific designs: a \textbf{Retrieval-enhanced Aggregator} that initializes generation using semantically similar warm items to bypass inefficient noise, and a \textbf{Simulation-based Representation Alignment} module that enforces distribution consistency between generated and real embeddings via contrastive learning. Experiments on three benchmarks confirm that DiffCold resolves the seesaw dilemma, consistently outperforming state-of-the-art methods across all metrics.

cs.IR↗
arXiv2026

LatentSkill: From In-Context Textual Skills to In-Weight Latent Skills for LLM Agents

Agent systems increasingly use textual skills to encode reusable task procedures, but injecting these skills into the prompt at every step incurs substantial context overhead and exposes skill content as plaintext. We present LatentSkill, a framework that converts textual skills into plug-and-play LoRA adapters through a pretrained hypernetwork. LatentSkill stores skill knowledge in weight space rather than context space, removing per-step skill tokens while preserving modular loading, scaling, and composition. On ALFWorld and Search-QA, LatentSkill outperforms the corresponding in-context skill baseline while using substantially fewer prefill tokens: it improves ALFWorld success by 21.4 and 13.4 points on the seen and unseen splits with 63.9% fewer prefill tokens on average, and improves Search-QA exact match by 3.0 points while using 71.8% fewer tokens per step. Further analysis shows that generated skill LoRAs form a structured semantic geometry, can be continuously modulated via the LoRA scaling coefficient, and can be composed through parameter-space arithmetic when skill components are aligned. These findings suggest that weight-space skills provide an efficient, modular, and less exposed substrate for extending LLM agents.

cs.CL↗
arXiv2026

Infinitesimal Bialgebra on Planar Binary Trees

We construct a weight-zero infinitesimal bialgebra structure on the $\bk$-module spanned by planar binary trees, using the under product $\backslash$ of Aguiar--Sottileand a root-recursive coproduct $\DeltaLR$. We prove that $\DeltaLR$ is coassociative and satisfies the infinitesimal derivation rule with respect to $\backslash$, hence gives a unitary infinitesimal bialgebra distinct from the usual Loday--Ronco Hopf structure. We also obtain an elementary vertex-cut formula, establish freeness properties for unitary $(\backslash,\vee)$-algebras and unitary infinitesimal $(\backslash,\vee)$-bialgebras, and identify the construction with the infinitesimal coproduct transported from planar rooted forests.

math.CO↗