Machine Learning
You Only Edit Once: Incentivizing In-Context Capability of LLMs via Local Demonstration Refinement
Jiarong Wen, Qi Wang, Yun Qu et al. (+8 more)
In-context learning (ICL) is crucial for boosting the inference performance of large
language models (LLMs). However, the effectiveness of ICL in LLMs is greatly influenced by the choice of demonstration sets. Exhaustive searches over these sets are combinatorial, and existing selectors often rely on relevance or likelihood proxies to implicitly assess ICL quality. Making repeated queries to the target LLM with these strategies can incur substantial costs. This work simplifies selection by framing it as a constrained local search problem and presents local demonstration editing (LDE). Starting with an initially retrieved set of demonstrations, LDE employs a single structured edit to explore its surrounding neighborhood while balancing performance gains with search costs. Technically, LDE is reduced to a policy search problem, for which we train a small LLM, referred to as Jev-LDE. This model as the System-1 modifies the retrieved demonstration set by performing actions such as Keep, Delete, or Replace elements, all within a framework of reinforcement learning with verifiable rewards. At test time, Jev-LDE executes a single edit of the retrieved demonstration set, followed by one inference from the target LLM, avoiding the need for iterative context scoring or subset searches. Across standard classification
benchmarks, various target LLMs with Jev-LDE as the plug-and-play module consistently improve ICL performance, and Jev-LDE shows transferability to held-out
benchmarks and models without retraining. These findings indicate that the LDE approach offers an efficient and adaptable method for harnessing the ICL capabilities of target LLMs.
Read the full abstract ▾
All authors Jiarong Wen, Qi Wang, Yun Qu, Yixiu Mao, Heming Zou, Haoang Chi, Lizhou Cai, Yiqin Lv, Kaiyu Zhang, Yuhang Jiang, Xiangyang Ji
Computational Engineering
Learning Transferable Reaction Mechanisms from Visual Chemical Knowledge
Yujian Yuan, Jiaxin Xu, Xin Cai et al. (+4 more)
Reaction mechanisms describe the step-by-step transformations underlying chemical reactions and are central to reaction analysis and synthesis. Learning-based models have achieved strong performance on established mechanism-prediction
benchmarks, but transferring them to unseen chemistry remains challenging. Such transfer is difficult because familiar mechanisms must be applied to unfamiliar molecular structures, and some target mechanisms may be poorly covered by the training data. To address these challenges, we introduce MechaVLM, a visual framework that combines transferable chemical representations with external mechanistic knowledge. It learns reusable visual features through multiscale chemical grounding and cross-rendering contrastive learning. For open-book prediction, MechaVLM retrieves a fixed set of precedents from 70,384 literature mechanism figures and re-reads relevant visual evidence as the molecular state evolves, directly using the figures without symbolic mechanism parsing. An atom-indexed
language decoder then recursively generates executable electron edits to construct the complete mechanism. We further introduce MechBench, a challenging literature-derived
benchmark with 2,184 mechanisms and 9,146 elementary steps. Across cross-
dataset and literature-derived
benchmarks, MechaVLM establishes strong zero-shot mechanism prediction. Its closed-book model alone improves Step/Pathway Top-1 by 12.50/13.93 percentage points on FlowER-to-ReactMech transfer, while external visual precedents unlock further gains on challenging OOD reactions. The learned representation also generalizes beyond mechanism prediction to atom mapping and reaction center prediction.
Read the full abstract ▾
All authors Yujian Yuan, Jiaxin Xu, Xin Cai, Yufan Chen, Zhichao Tan, Ziqi Zhou, Hanyu Gao
Machine Learning
Hierarchical Response Preservation for Continual Adaptation of Zero-Shot Graph-Text Models
Haopeng Zhang, Yuhan Wang, Yubing Su et al. (+4 more)
Pretrained graph-text models align graph representations with textual semantics, enabling recognition of unseen classes and transfer across graph domains. However, as graph data and classes continually arrive, models should learn from new supervision while retaining their zero-shot transfer capabilities and historical task knowledge. Two challenges arise: (i) new classes can overturn historical predictions despite preserved distinctions among historical classes, and (ii) overly strict response preservation can stall learning of new classes. To address these challenges, we propose Hierarchical Response Preservation (HiRP). HiRP represents this competition through a hierarchical response that keeps each historical-class probability and sums new-class probabilities, preserving historical distinctions and aggregate competition while allowing distinctions within the new class group to adapt. It further uses the geometry induced by this response to guide constrained updates, retaining useful adaptation directions while controlling response drift. Across three class-incremental settings, HiRP achieves absolute gains of 1.84-7.95 percentage points in average accuracy over the strongest compared baseline in each setting, while mitigating zero-shot transfer degradation.
Read the full abstract ▾
All authors Haopeng Zhang, Yuhan Wang, Yubing Su, Yingxin Chen, Xiao Wang, Ruijie Wang, Jianxin Li
Machine Learning
Short-Length Code Designs for Integrated Sensing and Communications: A Deep Learning Approach
Muah Kim, Shuangyang Li, Tayyebeh Jahani-Nezhad et al. (+2 more)
Integrated sensing and communication (
ISAC) enables joint communication and sensing using a shared waveform, but its signal design is challenging due to the inherent trade-off between the two objectives, particularly in the short blocklength regime. This paper proposes an autoencoder (AE)-based framework for
ISAC waveform design in noncoherent settings. We derive a modified Cramér-Rao bound for multi-target delay estimation and analyze the maximum-likelihood decoding rule for noncoherent communication under correlated fading. These results reveal structural connections and trade-offs between communication and sensing objectives in waveform design. Based on this analysis, the AE learns waveform representations that jointly optimize both functionalities, with a tunable parameter controlling the trade-off. Simulation results show that the proposed design outperforms conventional schemes in both communication reliability and sensing accuracy, especially under short blocklength and fading conditions.
Read the full abstract ▾
All authors Muah Kim, Shuangyang Li, Tayyebeh Jahani-Nezhad, Rafael F. Schaefer, Giuseppe Caire
Computer Vision
ViCoR: Reliable Molecular Structure Extraction via Spatially Aligned Verification and Executable Revision
Yujian Yuan, Xin Cai, Yufan Chen et al. (+5 more)
Reliable optical chemical structure recognition (OCSR) is essential for building high-quality chemical data from scientific literature, yet even small recognition errors can propagate into chemical databases and downstream models. In practice, recognized structures often require manual inspection and correction before use, making large-scale data curation costly and difficult to scale. We therefore study Selective Structure Recognition (SSR), a post-recognition setting that automatically produces reliable structured outputs while rejecting unresolved cases. Selection-only approaches can improve reliability by rejection, but cannot create additional correct outputs beyond those produced by the base recognizer. We propose ViCoR, a repair-before-rejection framework for iterative VerIfiCatiOn and Revision. Its key idea is to make observation-prediction correspondence explicit: coordinate-preserving rendering establishes spatial correspondence between the source image and predicted structure, while index anchoring maps localized visual discrepancies to executable graph edits without full-structure regeneration. A shared VLM is progressively trained from verification to revision. On two real-world OCSR
benchmarks, ViCoR improves overall accuracy from 73.53% to 88.26% and from 61.83% to 84.32%, while achieving over 97% accepted accuracy at 85–89% coverage. The resulting molecular data further improve reaction-extraction F1 by 15.5 points and literature-sourced reaction prediction accuracy by 7.7 and 5.8 points, demonstrating the value of automated reliability control for scientific data curation and downstream chemical learning.
Read the full abstract ▾
All authors Yujian Yuan, Xin Cai, Yufan Chen, Jiaxin Xu, Mengdi Liu, Zhichao Tan, Long Chen, Hanyu Gao
Artificial Intelligence
JustQuant: You Don't Need Smoothing, SVD, or Rotation for 4-Bit Activation Quantization
Kaicheng Yang, Kaisen Yang, Chunyu Liu et al. (+8 more)
Recent generative models have become increasingly powerful, but their inference cost continues to grow. Model quantization offers a promising way to compress these models and accelerate inference. However, at 4 bits, activation quantization is substantially more challenging than weight quantization. Recent post-training quantization (PTQ) and quantization-aware training (QAT) methods have made progress in 4-bit activation quantization by introducing smoothing, SVD branches, rotations, mixed precision, or advanced formats such as NVFP4. These additional operators and data types impose demanding requirements on inference engines and hardware, limiting the broad adoption of low-precision models. Can quantization be achieved using only plain low-bit operators? To answer this question, we propose JustQuant, a simple yet effective framework that moves the complexity of low-bit quantization from deployment-time operators into the training process. We first revisit model quantization from the perspective of knowledge
distillation and show that a key reason existing PTQ and QAT methods fail is that they typically exploit supervision at only a single level. We then introduce Theseus QAD, a quantization-aware
distillation method that progressively applies multi-level supervision, analogous to the gradual replacement process in the Ship of Theseus. Extensive experiments on DiT and
diffusion large
language models show two distinct regimes. For smaller models, Theseus QAD can serve as a lightweight warm-up stage that substantially improves subsequent QAT with plain operators, while naive QAD may collapse in the same setting. For larger models, Theseus QAD provides a stronger
distillation training path than ordinary QAD. Across both regimes, JustQuant improves low-bit quantization quality while avoiding the complex operators required by many existing PTQ methods.
Read the full abstract ▾
All authors Kaicheng Yang, Kaisen Yang, Chunyu Liu, Xianglong Yan, Haotong Qin, Junyi Wu, Tianao Zhang, Xun Zhang, Shaoqiu Zhang, Youbang Sun, Yulun Zhang
Graphics
Lattice Structure Optimization for Additive Manufacturing: Manufacturability-Driven Design and Pareto Front Construction
Yu Xing, Yang Liu and Lin Lu
Lattice metamaterials support lightweight, multifunctional structures, while additive manufacturing (AM) enables complex geometries. Yet multiphysics
lattice design faces two challenges: efficiently constructing well-covered multi-objective Pareto fronts under limited budgets, and satisfying manufacturing constraints such as overhangs, enclosed cavities, and restricted powder-removal channels. We propose a manufacturing-constraint-driven method for
lattice optimization and Pareto-front construction. Differentiable manufacturing constraints are embedded in inverse-homogenization
topology optimization, enabling joint optimization of physical performance and manufacturability. A progressive Pareto-front mechanism uses a density-generation network to learn latent representations of high-quality
lattices, interpolates neighboring nondominated representations, and decodes them into initial density fields for subsequent optimization. Newly found nondominated solutions update the network and sample set, progressively expanding the manufacturable set. On 3D periodic unit
cells, with 1000 optimization runs, network initialization achieves a 92.60% success rate and 916 manufacturable samples, versus 78.30% and 776 for random initialization. Its Pareto front reaches a hypervolume of 0.0787, compared with 0.0675 for random initialization. The results show that the method efficiently constructs broadly covered manufacturable Pareto fronts for multiple physical properties.
Read the full abstract ▾
Cryptography and Security
The Price of Peeking: Anytime-Valid Leakage Detection on ML-KEM EM Traces
Georgios Feretzakis and Alexandros Papaspyridis
Side-channel evaluators routinely inspect leakage tests while acquisition is still running, and extend or stop the campaign based on what they see. Fixed-horizon screening such as the Welch $t$-test with threshold $|t|>4.5$ gives no error guarantee for this monitored decision rule. We study anytime-valid leakage detection based on testing by betting: SKIT-type swap-pair e-processes whose false-alarm probability is controlled uniformly over time under an explicit conditional
symmetry null. In matched comparisons that share the frozen witness, rows and payoff, first-crossing detection needed 1.68-2.00$\times$ the traces of a fixed-horizon randomization test at 80% detection on synthetic streams, and 1.68-2.38$\times$ on degraded recordings from an open ML-KEM electromagnetic
dataset with the primary Ridge witness at $\alpha=0.05$. With the same primary witness and level, on undegraded reference and pqm4 recordings the monitored procedure stopped early: its median stopping point was 62-72 and 146-316 evaluation traces, i.e. 2-8% of a conservative 4096-trace budget. Under exact designed nulls on the recorded backgrounds, repeated-look $|t|>4.5$ screening over all 13000-20000 samples raised a false alarm in 2.7-12.9% of replicates, against 0.0-4.7% for terminal-only screening and no rejection by a sample-wise e-Bonferroni process, which in a prespecified follow-up detected natural-label associations in 4 of 4 backgrounds after 840-3288 traces. All recordings come from one device, and natural-label results are descriptive; we state the assumptions each claim requires.
Read the full abstract ▾
Robotics
Beyond One-Step Accuracy: State-Affine Latent Transition for Reliable Visual Planning
Boyuan Zhang, Yingjun Du, Xiantong Zhen et al. (+1 more)
Joint-
embedding world models enable visual planning by learning action-conditioned dynamics in latent space. Yet they are commonly trained for one-step prediction on encoded states, while planning recursively applies the learned transition to its own predictions. One-step accuracy therefore does not capture how prediction errors propagate under recursive rollout. We decompose multi-step rollout error into the errors introduced at individual steps and their propagation through subsequent transitions. We show that state-affine dynamics are precisely the differentiable transitions with state-independent Jacobians, eliminating the nonlinear propagation residual and making the error propagation operators depend only on the action sequence. Guided by this result, we introduce SALT (State-Affine Latent Transition), an action-conditioned state-affine dynamics model in which the action modulates both the state transformation and the additive update. We train SALT through recursive multi-step rollout supervision, feeding each predicted latent state back into the transition so that training matches how the model is used during planning. Across four visual planning environments, SALT exhibits $1.48$–$2.19\times$ higher one-step prediction error than the matched LeWM baseline, yet improves closed-loop success in every environment by $10.0$ percentage points on average. On OGBench-Cube, the fraction of episodes that fail with a sharp rise in model-predicted cost after execution decreases from $23.3%$ to $2.0%$.
Read the full abstract ▾
All authors Boyuan Zhang, Yingjun Du, Xiantong Zhen, Ling Shao
Computational Engineering
SymbolicLM: Training Language Models as Symbolic Regressors
Jun Yao, Yingfan Hua, Ruikun Li et al. (+4 more)
Large
Language Models (LLMs) have shown promising capabilities in scientific
reasoning, yet scientific discovery ultimately requires deriving precise laws directly from observational data, known as Symbolic Regression (SR). This poses a challenge for LLMs due to the gap between probabilistic text generation and the exact structural requirements of SR. Existing approaches rely on complex external scaffolds, which are computationally expensive and separate symbolic
reasoning from the model itself. To address this limitation, we propose to directly equip LLMs with symbolic regression capabilities through dedicated numerical-symbolic and physical supervision. We introduce PhysSymbArena, a large-scale
benchmark containing over 160,000 equations and 1.8B
tokens of numerical-symbolic data with physical descriptions, enabling systematic training and evaluation. Based on PhysSymbArena, we develop SymbolicLM, which enhances the symbolic regression ability of LLMs through mathematical and physical supervision. During inference, we further introduce SymbolicSGA, a refinement framework that leverages quantitative feedback to iteratively improve generated equations. Experiments on multiple symbolic regression
benchmarks show that SymbolicLM substantially improves structural recovery while maintaining competitive numerical fitting performance. These results demonstrate that symbolic regression can be explicitly learned as an intrinsic capability of LLMs.
Read the full abstract ▾
All authors Jun Yao, Yingfan Hua, Ruikun Li, Shixiang Tang, Bin Liu, Wanli Ouyang, Yan Lu
Computer Vision
LoopLUT: 3D Lookup Tables with Progressive Region Refinement for Real-Time 4K Image Enhancement
Yang Ye, Jiajun Ma, Chen Wu et al. (+5 more)
Color enhancement of 4K images must meet a quality target under a tight compute budget. Three-dimensional lookup tables (3D LUTs) dominate real-time enhancement because they decide at low resolution and apply a per-pixel lookup at full resolution. A single global LUT, however, is spatially invariant, so an underexposed shadow and a well-exposed region that share a pixel value receive identical corrections. Spatially heterogeneous demands cannot be expressed by such a mapping. We propose LoopLUT, a region-cascaded 3D LUT with progressive refinement. A global LUT performs the overall correction, followed by K-1 loop iterations. In each iteration a gating head predicts at low resolution the region that still needs correction, then builds a residual LUT from the color statistics of that region alone. The cascaded gates form a partition of unity, so the output is a per-pixel convex combination of the K lookup results. Fusion is therefore performed by the gates themselves, with no separate fusion module and no interpolation error accumulating across rounds. The decision stage runs at a fixed 256x256 resolution, independent of output resolution, so a 4K image costs only K pure lookups. Extensive experiments across four
benchmarks show that LoopLUT improves PSNR by up to 2.81 dB over the strongest prior method, while keeping real-time throughput at 4K. The same decomposition also generalizes well to
underwater enhancement
datasets.
Read the full abstract ▾
All authors Yang Ye, Jiajun Ma, Chen Wu, Wei Wang, Dianjie Lu, Guijuan Zhang, Linwei Fan, Zhuoran Zheng
Systems and Control
Control Data Scheduling over Shared Communication Channels: A Sparse and Collision-Free Mechanism
Yuxing Zhong, Zhaohua Yang, Fuhai Nan et al. (+2 more)
In systems where controllers operate remotely and communicate with actuators over shared communication channels, it is crucial to efficiently schedule the control data transmission to reduce bandwidth usage and actuator effort. In this article, we investigate a novel scheduling mechanism that jointly coordinates control actions across different controllers and different time steps. We propose an
algorithm based on the alternating direction method of multipliers (ADMM) to solve the resulting optimization problem. While ADMM is often treated as a black-box solver, the proposed
algorithm offers a clear physical interpretation, ensures convergence to a stationary point, and is computationally efficient. Simulation results validate our theoretical results and demonstrate the effectiveness of our proposed
algorithm.
Read the full abstract ▾
All authors Yuxing Zhong, Zhaohua Yang, Fuhai Nan, Daniel E. Quevedo, Ling Shi
Machine Learning
Shangzhen Zhu, Muyan Hu and Tomasz Kozlowski
Low-precision
Transformer systems increasingly quantize
attention matrix multiplications, while softmax often remains at higher precision. During pretraining, an approximate softmax changes the gradients that train the model as well as its forward computation. We study this interaction with K-interval
attention, which approximates the exponential using K+1 grid values. We vary per-row grid calibration, interpolation versus hard rounding, and the placement of a straight-through surrogate relative to normalization. We derive the corresponding backward rules, including calibration derivatives, and compare these choices in pretraining experiments matched on model, data, and optimizer. Detaching the row extrema leaves the forward computation unchanged but produces a delayed increase in validation loss. With hard rounding at K=4, min-max calibration and a pre-normalization surrogate incur a large loss gap; changing either choice substantially reduces it. At 124M parameters and 2.5B training
tokens, fixed-window calibration with a post-normalization surrogate yields a validation loss gap of +0.019 nats relative to softmax at K=4, and with a pre-normalization surrogate yields +0.004 nats at K=16.
Read the full abstract ▾
Machine Learning
TGRL: Temperature-Grouped Reinforcement Learning for Efficient Exploration in LLMs
Zihan Lin, Xiaohan Wang, Jie Cao et al. (+4 more)
Efficient exploration often remains a central bottleneck in reinforcement learning with verifiable rewards (RLVR). Although temperature control and test-time scaling strategies can increase rollout diversity of large
language models (LLMs), they either expand the sample budget at rollout time or leave the benefit of exploration unquantified. To this end, we propose Temperature-Grouped Reinforcement Learning (TGRL), which turns temperature-induced diversity into an explicit training signal. For each
prompt, TGRL partitions its rollout group into low- and high-temperature subsets, estimates exploration gain through their reward contrast, and allocates this group-level signal as
token-level credit using Jensen–Shannon (JS) divergence between the corresponding temperature-scaled next-
token distributions induced by the same logits. Notably, TGRL reaches equivalent accuracy up to 36% faster than strong RLVR baselines without expanding the rollout budget. Across 11
benchmarks from diverse domains, TGRL broadly improves over strong RLVR baselines: it improves the six-
benchmark math average by 1.6% at 32B, raises CodeForces rating by 196.7 points and LiveCodeBench Pass@16 by 4.4%, and improves ALFWorld/WebShop success rates by 6.3%/4.9%. Comprehensive ablations and wall-clock analysis confirm the efficacy of all proposed components. Code is available at https://github.com/1229095296/TGRL/tree/main.
Read the full abstract ▾
All authors Zihan Lin, Xiaohan Wang, Jie Cao, Jiajun Chai, Wei Lin, Guojun Yin, Ran He
Game Theory
On best of both worlds allocations with subadditive valuations
Uriel Feige
We consider allocation of indivisible goods to
agents with equal entitlements and subadditive valuations. As an ex-post
fairness notion we consider the maximin share (MMS), and as an ex-ante
fairness notion we consider the maximum expectation share (MES), which is always at least as large as the MMS, and sometimes much larger. We present a simple transformation that for every $0 < \rho \le 1$, given any
algorithm that produces $\rho$-MMS allocations, transforms it into a randomized allocation
algorithm that offers $\rho$-MMS ex-post simultaneously with $\eta$-MES ex-ante. We prove several new properties of MES, and use them to show that $\eta \ge \min[\frac{\rho}{2 + \rho}, \frac{1}{4}]$. We also present cases in which the transformation results in a higher value of $\eta$. Applying our transformation to currently known allocation
algorithms shows for subadditive valuations the existence of randomized allocations that are simultaneously $\Omega(\frac{1}{\log\log n})$-MES ex-ante and $\Omega(\frac{1}{\log\log n})$-MMS ex-post, and for XOS valuations the existence of randomized allocations that are simultaneously $\frac{4}{27}$-MES ex-ante and $\frac{4}{17}$-MMS ex-post.
Read the full abstract ▾
Machine Learning
Approximating Softmax in Pretrained LLMs: Model Sensitivity and Kernel Acceleration
Shangzhen Zhu, Muyan Hu and Tomasz Kozlowski
On NVIDIA Blackwell B200, tensor-core throughput outpaces special-function exponential throughput by more than two orders of magnitude, exposing exponential evaluation in fused
attention kernels. A pretrained
Transformer, however, may not need it evaluated accurately at every element. We characterize what a pretrained model does need by approximating softmax at inference in ten frozen decoder-only models (0.5B-72B). The number of positions the softmax map assigns probability to and within-row resolution can be cut substantially, yet uniform weighting of the same positions is damaging. Where a fixed resolution budget is placed matters as much as its size, with resolution near the row maximum consistently favored. Perturbations matched on scalar distortion produce model-dependent responses of opposite sign. These findings motivate Rowmax-PoT, a coarse logarithmic weight representation anchored at each row maximum, and Rowmax-H15, its hardware specialization in FlashAttention-4. On B200, the patched FP8
attention forward is 12.4% faster at causal 8K and 25.8% faster at non-causal 8K in host-side call-latency measurements; board energy per forward falls by 8.4% at causal 16K. Measured separately on the BF16 kernel path at 2K, Rowmax-H15 increases perplexity by 0.091-0.492% across five models from three families.
Read the full abstract ▾
Computer Vision
IVT-Guard: All-in-One Reasoning Model for AI-Generated Content Detection
Hongwei Niu, Yunpeng Luo, Hanjun Li et al. (+6 more)
The rapid proliferation of highly realistic AI-Generated Content (AIGC) necessitates robust and interpretable detection mechanisms. However, existing detectors are predominantly confined to single modalities and provide binary outputs without
reasoning. While
Multimodal Large
Language Models (MLLMs) present a promising solution, their development is constrained by the scarcity of
multimodal reasoning data and the
reasoning-detection optimization dilemma, where explicit
reasoning supervision can compromise detection accuracy. To this end, we introduce IVT-Set, a comprehensive
dataset comprising over 152K diverse image,
video, and text samples equipped with multi-granularity Chain-of-Thought (CoT)
reasoning trajectories. Based on it, we propose IVT-Guard, a pioneering framework for unified and interpretable AIGC detection across image,
video, and text modalities. Furthermore, to overcome the aforementioned optimization dilemma, we design a novel three-stage training paradigm: Artifact-Aware Pre-training, Artifact-to-Evidence Supervised Fine-Tuning via artifact-aware injection, and Evidence-Verdict Consistency Group Relative Policy Optimization. Extensive experiments demonstrate that IVT-Guard achieves state-of-the-art detection performance across in-domain, out-of-domain, and cross-
dataset settings while delivering faithful
reasoning. Code and data will be released.
Read the full abstract ▾
All authors Hongwei Niu, Yunpeng Luo, Hanjun Li, Ziyin Zhou, Jianghang Lin, Ke Yan, Shouhong Ding, Shengchuan Zhang, Liujuan Cao
Computer Vision
Fill2SR: Repurposing Inpainting Diffusion Transformers for Real-World Super-Resolution
Xingfu Yi and Xiaoxue Yu
Recent real-world image super-resolution (SR) methods often adapt text-to-image (T2I) backbones with ControlNet-style branches or spatial conditioning
tokens, which increases
memory and computes with resolution and often constrains training to a fixed scale. We propose Fill2SR, which repurposes a masked-inpainting
Diffusion Transformer for SR without extra spatial branches. Our Inpainting-Interface Evidence Adapter (IIEA) writes the low-quality (LQ) observation into the native masked-image slot under a full-image mask, turning inpainting into a reverse-degradation conditional rectified flow trained with LoRA-only tuning. We further introduce RCDT, an offline pipeline that distills degradation descriptors from unpaired real images and transfers them onto clean targets using frozen open-source models. Fill2SR supports mixed-resolution training up to QHD and yields stable performance across $512/1024/2048$ outputs. On synthetic
benchmarks, our base model with IIEA achieves the best LPIPS on DIV2K and LSDIR; adding RCDT trades a small LPIPS drop for consistently stronger no-reference quality on RealLQ250 and RealPhoto60. Fill2SR remains
memory-predictable, running $1536^2$ inference on a single 32GB GPU and extending to multi-megapixel outputs via tiled restoration.
Read the full abstract ▾
Computer Vision
ForeFly: A Dual-Horizon World Action Model for Aerial Vision-Language Navigation
Kunhui Wang, Xintong Zhang, Junyu Gao et al. (+1 more)
Aerial
Vision-
Language Navigation (AVLN) requires
UAVs to maintain reliable instruction following over long trajectories in complex 3D environments. However, existing AVLN approaches are predominantly reactive or limited to single-horizon prediction, overlooking complementary future cues across different temporal horizons. To address this limitation, we propose ForeFly, a dual-horizon latent world action model that predicts both a proximal future for local continuity and an adaptive route-critical future for long-range guidance. Horizon-specific foresight queries are primed with recent and route-critical visual
memories, providing history-aware context for future prediction. To exploit their distinct roles in action generation, we introduce Foresight-Guided Action Refinement (FGAR), which asymmetrically exploits proximal foresight for local action enhancement and route-critical foresight for feature-wise correction and route-level guidance. Experiments on the TravelUAV and
UAV-ON
benchmarks show that ForeFly consistently outperforms strong baselines across seen and unseen settings, validating the effectiveness of dual-horizon foresight and FGAR learning. The code is available at: https://github.com/kunhuiW/ForeFly
Read the full abstract ▾
All authors Kunhui Wang, Xintong Zhang, Junyu Gao, Changsheng Xu
Machine Learning
Compressing Value Predictions for Learning-Augmented Metrical Task Systems
Sizhe Li, Yecheng Li and Kun He
Learning-augmented
algorithms for metrical task systems (MTS) can exploit predictions of canonical dual values, but existing formulations typically require a prediction for every state. We study whether these predictions can be compressed to a small set of representative states while retaining their algorithmic value. We introduce landmark-compressed value predictions, in which the predictor reports predicted dual values only at $m$ landmarks and the remaining values are reconstructed by a Lipschitz extension. Our
algorithm achieves additive excess cost $O(T\,r(L) + \sum_t \delta_t)$, where $r(L)$ is the covering radius of the landmarks and $\delta_t$ measures prediction error up to additive shifts; local and value-dependent bounds refine this guarantee. For sparse landmark sets on unit-spaced finite lines, we prove a matching $\Omega(T r_m)$ lower bound for every randomized
algorithm using fixed landmarks, even with advance access to their entire exact absolute-value table. The prediction interface also matters: on two states with one landmark, exact absolute values permit horizon-independent excess, whereas exact relative values force worst-case expected excess linear in $T$. We give PAC guarantees for learning compressed prediction tables, with efficient empirical-risk minimization for fixed landmarks. Our results connect metric coverage, prediction interfaces, and learning guarantees for compressed predictions in online MTS.
Read the full abstract ▾
Artificial Intelligence
OpenFC: Learning Verification Policies towards Open-Search Fact Checking
Xinming Wang, Kaixiang Qiu, Yansong Lin et al. (+5 more)
Open-search fact checking is not merely
retrieval followed by classification, but a sequential decision problem in which every query, source visit, and stopping decision reshapes the evidence available for verification. Yet existing systems often distribute these decisions across predefined pipelines or separately prompted modules rather than learning them as a unified task-specific policy. We introduce OpenFC, a unified verification-policy training framework that post-trains Qwen3-8B as a compact next-action controller over
reasoning, evidence acquisition, and stopping. OpenFC learns this policy in two stages. Stepwise-Calibrated Cold Start (SCCS) uses a strong training-time supervisor to review post-initial
reasoning, tool-use, and stopping proposals before execution, producing reliable trajectories for supervised fine-tuning without access to gold verdicts. Verification-Aware Reinforcement Learning (VA-RL) then improves the cold-start policy on unresolved claims through budget-aware tool rewards, label-aware advantage reweighting, and localized response masking. Across six fact-checking
benchmarks, OpenFC achieves 70.39% average accuracy and 63.30% macro-F1, the highest overall averages among the evaluated methods. Stage-wise ablations further show that SCCS and VA-RL provide complementary gains, supporting the design of the two-stage training framework. These results position OpenFC as a strong and effective framework for open-search fact-checking. We will open-source our code and release the model checkpoints to support reproducibility.
Read the full abstract ▾
All authors Xinming Wang, Kaixiang Qiu, Yansong Lin, Chunji Lv, Yi Chen, Boran Wang, Hong-Ming Yang, Xu-Yao Zhang
Robotics
SLIP-VLA: Single-Step Latent Imagination for Policy Learning in Vision-Language-Action Models
Tianfu Li, Haoxuan Xu, Wenbo Chen et al. (+7 more)
Vision-
Language-Action models are increasingly effective for robotic manipulation, yet most predict actions directly from current observations without explicitly modeling future scene
evolution. Recent methods introduce future prediction to improve action generation, but dense future modeling often requires expensive iterative denoising, while one-step alternatives can underperform their multi-step counterparts. To reconcile efficient future modeling with strong action performance, we present SLIP-VLA, a policy learning framework that equips VLA models with a Single-Step Latent Imagination for future-aware action prediction. SLIP-VLA obtains temporally dense future latent representations with a single denoising update, and we improve the perceptual sufficiency of these representations by aligning intermediate latents with future
geometric and semantic features. We further improve their control sufficiency through action-conditioned latent world modeling and inverse dynamics modeling, explicitly coupling latent transitions with
robot actions. SLIP-VLA achieves state-of-the-art performance across diverse simulation
benchmarks and real-world manipulation tasks, while its single-step latent imagination takes only 12 ms.
Read the full abstract ▾
All authors Tianfu Li, Haoxuan Xu, Wenbo Chen, Haitian Li, Changchuan Yang, Xinhu Zheng, Jun Ma, Yuan Liu, Lujia Wang, Haoang Li
Machine Learning
HiLoRe: What to Store, Compress, or Recompute for Efficient GRPO Training
Xinrui Chen, Mengyang Li, Ou Wu et al. (+1 more)
Group-relative policy optimization (GRPO) makes learner-side activations a major
memory-computation bottleneck: gradient checkpointing reduces activation
memory through recomputation, but fixed schedules can leave roughly 18 GB unused on a 48-GB GPU despite substantial recomputation overhead. Existing activation-management methods set state fidelity from execution cost, tensor properties, or generic compression sensitivity, without explicitly incorporating GRPO's analytic update structure into state-fidelity allocation. We formalize this dependence as policy-update exposure, linking the current GRPO loss coefficients to state-level approximation sensitivity. These coefficients are available before backward without an additional backward pass. We introduce HiLoRe, which allocates graph-attributed recovery units among high-precision storage, low-precision compression, and deterministic recomputation using measured recovery utility and update-conditioned approximation risk. It combines high-precision storage and deterministic recomputation with low-precision recovery under a calibrated risk budget. Across five model-task settings with 2K responses and
memory < 1.10 times GC's per-GPU actor-update peak, HiLoRe's actor-update throughput gains reach 13.5% over GC and 7.9% over the fastest evaluated baseline, with paired mean downstream-score differences below 0.6 percentage points.
Read the full abstract ▾
All authors Xinrui Chen, Mengyang Li, Ou Wu, Ji Zhang
Cryptography and Security
Information Blackhole: Exploring Backdoor Mechanism in 3D Point Cloud Reconstruction
Zhifei Yang, Xiuping Liu, Kuofeng Gao et al. (+3 more)
Point cloud autoencoders are fundamental components for 3D world representation and support many safety-critical downstream applications. Existing studies have extensively investigated
backdoor attacks on point cloud classification, whereas
backdoor attacks against point cloud autoencoders remain largely unexplored. However, their
backdoor behaviors differ substantially due to the intrinsic structural gap between discriminative and generative models. Specifically, a classifier is a discriminative model that separately fits the marginal distributions of benign and malicious data. In contrast, the generative nature of an autoencoder entangles the two within a unified latent distribution, leading to information crosstalk and reduced attack controllability. In this setting, residual source
geometric information in malicious data may leak into the clean inference branch, causing the reconstruction to collapse toward the source data. We then propose the Information Blackhole principle, which introduces Gaussian distribution constraints to disentangle latent representations and block interfering information. Building on this principle, we further propose Adaptive Gaussian Matching (AGM), which explicitly regularizes the latent distribution of poisoned samples. By suppressing the propagation of source
geometric information from poisoned features to the attacker-specified reconstruction target, AGM improves attack controllability. Extensive quantitative and qualitative experiments on ModelNet and ShapeNetPart demonstrate that the proposed framework improves the attack performance of several standard triggers and reveals the unique operating mechanisms of
backdoor attacks against point cloud autoencoders.
Read the full abstract ▾
All authors Zhifei Yang, Xiuping Liu, Kuofeng Gao, Junkai Qiu, Meng Liu, Yuhao Bian
Machine Learning · Cross-listed from quant-ph
Terminal-Register Certification for Finite-Measurement Learning of Multiscale Quantum States
Bhvain Makwana, Kashyap Patel, Manjunath Joshi et al. (+1 more)
Structured
quantum-state learning not only depends on an expressive ansatz but also on an operational certificate that stays meaningful with finite measurements and imperfect implementation. We study pure one dimensional states learning by an inverse binary multiscale
entanglement renormalization ansatz (MERA). In the learning procedure, the
qubits removed during coarse graining are controlled coherently and measured together at the terminal register. We confirm that an ideal sequential and terminal measurement schedule delivers the same complete bit string distribution under matched causal operations, while normalized postselection can amplify perturbations inversely with prefix acceptance. A noise aware
theorem introduces an individual calibrated total variation implementation budget to the finite shot certificate. The protocol is estimated on an open boundary transverse field Ising ground state. A frozen 8-
qubit schedule using $560$ million simulated training measurements per run achieves fidelity above $0.99$ in all $60$ held-out runs, with a mean fidelity of $0.996886$. 1080
circuit-noise
cells and 6480 confidence-coverage rows are covered by fixed-
circuit robustness validation without a locked soundness violation. We then address architectural
fairness at $n=16$ using three new studies. In a 120-run exact-gradient multistart diagnostic, MERA has higher fidelity in 58/60 paired restarts and lower long-range error in 60/60, although no run met the prespecified stationarity criterion. Finally, a causal cone-complete, parameter matched local
circuit achieves $2.62\times$ greater aggregate gate exposure yet loses all 30 paired comparisons in fidelity, long-range error, energy, and
entropy.
Read the full abstract ▾
All authors Bhvain Makwana, Kashyap Patel, Manjunath Joshi, Jaideep Mulherkar
Machine Learning
Correct then Forecast: Observer State-Space Models for Time Series Forecasting
Alexis-Raja Brachet, Guillaume Clavier–Frémond, Abdelhakim Ziani et al. (+2 more)
Time series
forecasting requires extrapolating the dynamics of an observed process beyond the last available measurement. Yet recurrent
forecasting models typically treat observations as inputs that directly control their latent dynamics. It leads to a regime change when these observations become unavailable at prediction time. Following a state-estimation perspective, we introduce Observer State-Space Models (OSSMs), a class of recurrent models that interprets the observed input time series as measurements of an underlying autonomous dynamical system. OSSMs explicitly separate latent-state propagation from measurement assimilation: a single transition governs the dynamics across both context and
forecasting intervals, while available observations correct the estimated state through an observer. This formulation naturally exposes classical control-theoretic properties, including observability and convergence of the state estimation error. We further show that conventional and recent SSMs can be recovered as particular instances of our OSSM framework, thereby providing a unified interpretation of their recurrent dynamics and revealing modeling inconsistencies. We perform experiments across several
benchmarks showing that OSSM achieves substantial improvements while maintaining the same parameter count and training setup as the corresponding SSM baseline. These results support a simple principle for recurrent
forecasting: observations should correct the estimated latent state, rather than control the dynamics used to propagate it.
Read the full abstract ▾
All authors Alexis-Raja Brachet, Guillaume Clavier–Frémond, Abdelhakim Ziani, Pierre-Yves Richard, Céline Hudelot
Artificial Intelligence
Dr. Free: You Don't Need Difficulty Rewards for Self-Evolving Search Agents
Zhipeng Qian, Zihan Liang, Yufei Ma et al. (+10 more)
A central limitation of current data-free self-
evolution methods for training search
agents is their reliance on difficulty-based proposer rewards. These methods reward a proposer for generating questions that challenge a co-evolving solver, using solver difficulty as a proxy for question quality. Yet difficulty alone is insufficient to distinguish questions that require cross-passage evidence from those that are answerable via simpler shortcuts. In addition, measuring difficulty demands repeated solver rollouts for every candidate question, leading to substantial computational costs. In this paper, we introduce \methodname, the first self-evolving search framework that eliminates difficulty-based proposer rewards and directly optimizes for evidence necessity relative to shortcut contexts. Dr. Free samples relational chains from a knowledge graph and pairs them with aligned passages, giving question generation an explicit multi-hop structure. A generated question receives a positive information-gain reward only when the likelihood of the target answer under the complete evidence passages exceeds the maximum likelihood under all evaluated shortcut contexts. Because this signal is computed from teacher-forced likelihoods, it removes the need for pass-rate estimation and reduces proposer training time by over $7\times$. Experiments on seven open-domain QA
benchmarks show that Dr. Free outperforms prior data-free search
agents and the supervised baseline, with large improvements on multi-hop QA
benchmarks.
Read the full abstract ▾
All authors Zhipeng Qian, Zihan Liang, Yufei Ma, Jie Ma, Ben Chen, Huangyu Dai, Lingtao Mao, Xinyu Sun, Tong zhao, Xuxin Zhang, Qingpeng Cai, Peng Jiang, Qibin Hou
Machine Learning
MA-JEPA: Joint-Embedding World Models for Multi-Agent Reinforcement Learning
Brandon Gary Kaplowitz, Osaze James Obahor and Christian Schroeder de Witt
World models improve sample efficiency by training policies on imagined trajectories, but their usefulness depends on learning representations that capture the information needed for future control. We study whether self-supervised joint-
embedding prediction (JEPA) can provide this learning signal for multi-
agent reinforcement learning. We introduce MA-JEPA, a
stochastic world model that replaces observation reconstruction with prediction of target representations, enabling model-based multi-
agent reinforcement learning with centralized training and decentralized execution. A categorical latent state and a causal
Transformer are trained with posterior and action-conditioned dynamics prediction objectives and are then used for actor-critic learning from latent imagination. A training-only joint predictor conditions on all
agents' local states and actions to predict each
agent's next local observation
embedding. These predictions are passed through the same local posterior used during real interaction with a centralized critic that is used only for value learning, with execution remaining decentralized. Our experiments show that this architecture performs strongly on SMAC, matching or exceeding the strongest reported comparator mean win rate on four of eight evaluated maps.
Read the full abstract ▾
Machine Learning
OOD Generalization as a Bifurcation Problem
Nguyen-Thanh-Luong Doan, Quang-Vu Nguyen, Tang-Phu-Quy Le et al. (+1 more)
Systematic out-of-distribution (OOD) generation remains a critical bottleneck for continuous-time generative models. While standard joint classifier-free guidance (CFG) routinely fails to synthesize unobserved concept combinations, exact decomposed scoring generalizes robustly at the cost of severe computational overhead. In this work, we reveal that compositional binding is not a uniform process but a highly localized phase transition. We identify the semantic bifurcation window - the precise temporal interval where joint and decomposed vector fields meaningfully diverge. Exploiting this dynamic, we propose
surgical guidance, a hybrid sampling strategy that restricts exact multi-pass scoring strictly to this critical window. On an OOD bi-digit MNIST testbed,
surgical guidance achieves state-of-the-art compositional fidelity at a fraction of the inference cost, yielding a +5.3% absolute improvement in pairwise accuracy over the joint baseline by intervening during just the first 15% of the
diffusion trajectory. Furthermore, our empirical analysis uncovers a fundamental topological divide:
diffusion models (SDEs) force conceptual resolution immediately at peak noise, whereas Conditional Flow Matching (ODEs) delays structural binding until intermediate features emerge, establishing a new temporal framework for accelerating large-scale generative decoding.
Read the full abstract ▾
All authors Nguyen-Thanh-Luong Doan, Quang-Vu Nguyen, Tang-Phu-Quy Le, Cong-Phap Huynh
Machine Learning
GraphSelect for Budgeted Representation Selection in Multimodal Graph Inference
Xu Wang, Xunkai Li, Yinlin Zhu et al. (+1 more)
Multimodal graph predictors combine text, images, and relations to classify connected entities. How much of this input is needed to preserve their predictions? We study budgeted representation selection, which chooses a subset of candidate text and image vectors under a separate capacity for each modality. Predictions from the complete candidate input define the classes to preserve. The challenge is that a representation's contribution depends on the other selected inputs, while graph propagation extends its effects across nodes. Our empirical study shows that candidate rankings change with the selected input, while predicted probabilities remain informative after the class stops changing. Updating scores improves selection, and exchanging inputs can improve a subset whose capacity is already filled. These findings lead to GraphSelect, which starts from individual candidate gains and refines the subset through jointly evaluated exchanges. It screens promising removals and additions, accepts an exchange when it reduces the prediction loss, and updates the scores. Experiments on six graphs show higher mean objective recovery than six attribution and explanation methods adapted to the selection task. Across nine trained architectures on two graphs, retaining 20% of the candidate representations per modality gives a mean accuracy drop of 0.10 percentage points relative to full candidate input, preserving classification performance with substantially fewer text and image representations.
Read the full abstract ▾
All authors Xu Wang, Xunkai Li, Yinlin Zhu, Rong-Hua Li