arXiv:2606.04268v1 Announce Type: new
Abstract: LZ77-based codecs exhibit a fundamental sequential bottleneck in decoding: each back-reference depends on previously decompressed data, preventing multi-core scaling. We present ACEAPEX, a parallel LZ77 codec that stores all back-references as absolute positions in the decompressed output and organizes data into self-contained 1 MB blocks, enabling embarrassingly parallel block-level decoding. Integrated into lzbench, ACEAPEX achieves 10,160 MB/s on EPYC 4344P (8 cores) and 10,869 MB/s on EPYC 9575F for FASTQ genomic data -- up to 3.1x faster than zstd -3 at comparable compression ratios. We further implement a GPU wavefront decoder on NVIDIA H100 SXM, measuring 44.0 GB/s on enwik9 and 20.3 GB/s on FASTQ (wavefront match phase, BIT-PERFECT verified). With a depth-limited encoder variant (-1.5% ratio on enwik9), GPU throughput reaches 77.2 GB/s on a single H100 and 249.9 GB/s on two H100s in NVLink configuration. To our knowledge, this is the first reported GPU LZ77 decode with near-standard compression ratio verified byte-for-byte.
Science Journals
arXiv:2606.05001v1 Announce Type: new
Abstract: With the telecommunications field embracing zero touch management alongside novel O-RAN and AI-RAN frameworks, contemporary telecom networks now function as immensely intricate and heavily softwareized codebases. While automated software engineering (ASE) tools and Software Engineering (SWE) Agents hold the potential to alleviate the critical code generation bottleneck in this domain, their ability to navigate and modify specialized, mathematically rigorous wireless stacks like srsRAN 5G remains unverified. General-purpose coding benchmarks fail to capture the stateful logic and strict requirements of telecommunications, leaving a critical evaluation gap. In this paper, we introduce TeleSWEBench, the first commit-driven benchmark specifically designed to measure an agent's performance in the telecom domain. We mine real developer commits from the srsRAN 5G repository and distill them into structured test cases across three difficulty tiers (Easy, Medium, and Difficult). Our benchmark consists of 734 questions that are accompanied by executable unit tests. To avoid the rigidity of test cases, we further propose a hierarchical LLM as a Judge framework called TeleJudge that scores agent outputs at the file level and aggregates verdicts holistically. This follows an evaluation based on context and semantic similarity in parallel to a standard unit test-based evaluation. Using this benchmark, we evaluate AIDER, OpenHands, and the ClaudeCode frameworks, powered by state-of-the-art reasoning LLMs, including Qwen3, GPT OSS, Gemma 4, Kimi, and Qwencoder 2.5. Our two-stage evaluation reveals that models suffer from a lack of both localization accuracy and functional correctness, with the strongest ASE tools achieving up to 25% of shippable changes.
arXiv:2606.04543v1 Announce Type: new
Abstract: Artificial intelligence in education is evolving from passive chatbots to proactive AI agents capable of initiation and goal-directed interactions. While offering opportunities for personalised learning, this shift risks undermining learner agency and cognitive effort. This paper reviews six pedagogical principles-prior knowledge activation, collaborative learning, problem-based learning, formative assessment, scaffolding, and metacognition-through the lens of agentic AI. We discuss the tension between automation and learning, proposing design recommendations that prioritise intentional friction, dynamic scaffolding, human-in-the-loop oversight, and considered AI utilisation to ensure AI supports rather than supplants human learning.
arXiv:2606.04877v1 Announce Type: new
Abstract: Proof assistants based on expressive logics suffer limited automation for proof search, raising the cost of formal verification based on proof assistants. We address this problem by introducing the Abduction Prover for Isabelle/HOL. Given a challenging proof goal, the Abduction Prover constructs a proof script for the goal by identifying useful conjectures using abductive reasoning.
arXiv:2603.04444v4 Announce Type: replace
Abstract: As large language models (LLMs) diversify across modalities, capabilities, and cost profiles, the problem of intelligent request routing: selecting the right model for each query at inference time, has become a critical systems challenge. We present vLLM Semantic Router, a signal-driven decision routing framework for Mixture-of-Modality (MoM) model deployments. The architecture follows two complementary Shannon-inspired views. In the information-theoretic regime, signal extraction reduces the entropy of "which model?" by distilling routing-relevant information from raw queries. In the Boolean-algebraic regime, the decision engine composes functionally complete routing policies from signal conditions. The central innovation is composable signal orchestration: thirteen heterogeneous signal types, spanning sub-millisecond heuristics and neural classifiers for semantics, safety, and modality, are composed through configurable Boolean decision rules into deployment-specific routing policies, so that fundamentally different scenarios (multi-cloud enterprise, privacy-regulated, cost-optimized) are expressed as different configurations over the same architecture. Matched decisions drive semantic model routing via thirteen selection algorithms, while per-decision plugin chains enforce safety constraints including a three-stage HaluGate hallucination detection pipeline and a lightweight episodic memory system with ReflectionGate for personalized multi-turn context. A typed neural-symbolic DSL specifies these routing policies and compiles them to multiple deployment targets, enabling configuration-first adaptation without code changes. Together, these components show that composable signal orchestration enables a single framework to serve diverse deployment scenarios with differentiated cost, privacy, and safety policies.
arXiv:2606.03784v2 Announce Type: replace
Abstract: Embodied chain-of-thought (CoT) aims to bridge linguistic reasoning and robotic control, but its effective form and integration strategy remain underexplored. In this paper, we revisit embodied CoT for vision-language-action (VLA) models at large scale. We construct the largest embodied CoT corpus to date, comprising 978,743 trajectories, 226.3M samples, and 2592.5 hours of robot data. Through extensive experiments, we find that effective embodied CoT should ground high-level semantic understanding into concrete action guidance, such as end-effector movement descriptions and image-space trajectories, while high-level reasoning alone brings only marginal gains. We further show that explicit CoT does not scale reliably when used as an autoregressive action prefix, as it suffers from compounding inference errors and unstable reasoning-action coupling. To address these limitations, we propose ERVLA, a VLA model that uses embodied CoT as representation-shaping supervision rather than mandatory test-time reasoning. ERVLA is trained with a reasoning-dropout strategy, enabling the model to absorb rich reasoning traces during training while predicting actions directly without CoT decoding during inference. This design improves scalability with increasing pre-training data and avoids autoregressive instability. ERVLA achieves state-of-the-art performance on LIBERO-Plus with an 86.9% success rate and reaches 53.2% success rate on VLABench, demonstrating strong out-of-distribution generalization. In real-robot experiments, ERVLA further outperforms competitive state-of-the-art baselines, especially on tasks requiring semantic disambiguation and long-horizon execution.
arXiv:2606.04775v1 Announce Type: new
Abstract: Text-to-video (T2V) models trained on large-scale web data can generate undesired content, motivating interventions that reduce harmful outputs without sacrificing visual quality. Activation steering offers an attractive mechanistic alternative to finetuning and prompt filtering, but existing T2V steering methods remain limited, typically applying coarse, non-anticipative interventions that can lead to oversteering and content degradation. To close this gap, we propose Latent Activation Linear-Quadratic Regulator (LA-LQR), a reduced-order optimal control framework for minimally invasive T2V steering. LA-LQR formulates T2V inference as a dynamical system and computes closed-loop feedback interventions that steer activations toward desired feature setpoints while penalizing unnecessary perturbations. To make optimal control feasible for high-dimensional video activations, we project activations onto a low-dimensional, task-relevant subspace derived from contrastive prompt pairs, estimate local linear dynamics in this latent space, and solve a latent LQR problem to obtain timestep- and layer-specific steering signals. We provide theoretical bounds relating latent setpoint tracking to raw activation-space feature control, and empirically validate the fidelity of the reduced latent dynamics. On concept steering and video safety benchmarks, LA-LQR reduces unsafe generations relative to baselines, while preserving prompt fidelity and visual quality.
arXiv:2411.19758v2 Announce Type: replace
Abstract: Remote sensing change detection based on a map reference and an up-to-date image boosts timely observation of the Earth's surface when earlier images are lacking for comparison. However, the semantic gap between high-level map categories and low-level image details hinders the extraction of homogeneous features for robust temporal association in change detection. Unlike conventional approaches that either compare pixel-level visual similarity or propagate segmentation errors, \textcolor{black}{we propose a novel framework, \underline{La}nguage-\underline{VI}sion \underline{D}iscriminator for d\underline{E}tecting changes, LaVIDE}, which bridges the semantic gap between high-level map categories and low-level image details using language as an intermediary. Specifically, we introduce {\it restricted prompt learning} to generate context-aware textual prompts that align map semantics with image content, and an {\it object-aware embedding enhancement} strategy to integrate object-level attributes (e.g., shape, boundary) into map representations. These components enable robust cross-modal alignment within a unified language-vision feature space. Extensive experiments on four benchmarks, DynamicEarthNet, HRSCD, BANDON, and SECOND, demonstrate that LaVIDE outperforms state-of-the-art methods by significant margins, achieving $18.4\%$ and $5.2\%$ improvements in IoU on multi-class and single-class change detection tasks, respectively. Our framework not only advances the accuracy of map-image change detection but also provides a practical solution for rapid map updating with minimal human intervention, promising broad impacts in urban planning, disaster assessment, and ecological conservation. Code and datasets are available at: https://github.com/ShuGuoJ/LAVIDE.git.
arXiv:2606.04429v1 Announce Type: cross
Abstract: A common heuristic used to explain the generalization of first-order gradient methods on non-convex neural networks is that "flat interpolators generalize well" (Hochreiter and Schmidhuber, 1994; Keskar et al., 2017), where flatness can be measured by the trace of the Hessian of the empirical loss. However, Dinh et al. 2017) showed that, using symmetry of the network that can change flatness while keeping the population and empirical losses unchanged, any interpolator can be made sharper or flatter. This result makes the earlier heuristic statement vacuous. In this paper, we show that for learning an unknown multi-index model with $2$-layer non-convex homogeneous neural networks, there is a connection between flatness and generalization, despite the existence of symmetries. This connection pertains to the "flattest" interpolators, i.e., the interpolators that have orderwise minimum flatness among all interpolators. First, we show that there exists a natural class of non-generalizing interpolators whose flatness cannot be made closer to the flattest possible, even using symmetries. Second, we show that for data generated by a sum of single-index models, if the approximation error and label noise are low, any flattest interpolator achieves small population loss, i.e., the flattest interpolators always generalize. This establishes a direct link between flatness and generalization which applies to a large class of activations and realistic data distributions.
arXiv:2606.04155v1 Announce Type: new
Abstract: Social skills such as negotiation and leadership are crucial for personal and professional success in today's interconnected world. However, scalable and effective training remains a significant challenge due to the scarcity of expert coaching. In this paper, we introduce SocialCoach, a holistic LLM-powered agentic tutoring system for personalized social skill development at scale. First, SocialCoach automatically constructs a pedagogically-grounded, theory-to-practice knowledge corpus from diverse expert sources, leveraging a multi-agent pipeline. Second, to personalize the learning journey, it employs an adaptive practice scheduling module that follows a prescription-retrieval-adaptation process. To maximize the long-term learning experience while overcoming the cold-start problem, this policy is optimized within a learner simulation environment through reinforcement learning. Finally, SocialCoach integrates immersive, goal-driven practice, causality-driven proficiency assessment and knowledge-grounded, reflective tutoring to help address the knowing-doing gap. We deploy it in our product, EQoach, and conduct extensive experiments. The results show that SocialCoach improves simulated pathway quality and judge-rated tutoring quality over baseline approaches, while early user feedback indicates strong perceived engagement and usefulness. These findings suggest a practical architecture for personalized and gamified pedagogical platforms on soft skill learning.
arXiv:2606.04737v1 Announce Type: new
Abstract: Large-scale video generation models have made remarkable progress in semantic consistency and visual quality, producing videos that are increasingly coherent and visually convincing. Nevertheless, the dynamics induced by pixel-level fitting do not naturally accommodate the regularities that govern real-world motion and interaction, resulting in persistent shortcomings in physical plausibility. To address this limitation, we propose \textbf{PILA} (Physics-Informed Latent Alignment), a framework that injects physics-structured latent guidance into the frozen flow-matching dynamics of pretrained video models. Specifically, PILA first employs anchored field estimation to map frozen-generator latents into an operational physical attribute bank organized by field-proxy slots, using observable motion as a kinematic anchor for constructing less directly observed proxies. To handle the heterogeneity of real-world dynamics, PILA adopts a mixture-of-experts design over physical categories. Label-prior masked expert routing selects category-specific operator experts, whose refinements are regularized by operational residuals abstracted from physical relations. Finally, the refined proxies are fused into the physical attribute bank and decoded into a correction to the flow-matching vector field, injecting physics-aware guidance while preserving the visual prior of the pretrained backbone. With staged adapter training on Wan 2.1-1.3B and direct transfer of the learned adapter to Wan 2.2-14B, PILA achieves state-of-the-art results on VBench-2.0, VideoPhy-2, and PhyGenBench in both visual quality and benchmark-measured physical plausibility.
arXiv:2606.04419v1 Announce Type: cross
Abstract: MRI provides excellent soft-tissue contrast without ionizing radiation, but long acquisition times increase patient discomfort while also raising exam costs and limiting scanner throughput. A common approach to reduce scan time is to acquire fewer measurements, which yields an ill-posed linear inverse problem; recovering diagnostic-quality images therefore requires incorporating prior knowledge beyond the measured data. In follow-up exams, the most recent prior scan of a patient can provide a highly informative subject-specific context, but practical use is complicated by temporal changes (including pathology progression), misalignment between scans, and protocol drift across acquisitions. In this work, we introduce L-TGVN, a Longitudinal Trust-Guided Variational Network that leverages prior scans as side information to reconstruct the current scan from heavily undersampled measurements. Crucially, L-TGVN constrains the influence of prior scans to be consistent with the acquired measurements. Unlike many existing longitudinal reconstruction methods, it does not require explicit pre-registration between prior and current scans. It further accommodates differences in acquisition protocols across visits (e.g., changes in sequence parameters). We evaluate L-TGVN against matched-capacity baselines, including prior-guided methods and methods that do not use longitudinal priors, and observe consistent improvements in standard quantitative metrics together with better preservation of fine structures at challenging accelerations. Source code is available at github.com/sodicksonlab/L-TGVN.
arXiv:2606.04376v1 Announce Type: cross
Abstract: Real-time multi-camera 3D reconstruction is a key foundation for immersive media, remote interaction and spatial computing. While synchronized camera arrays are widely adopted, achieving geometrically consistent and scalable real-time reconstruction remains challenging. A key challenge is the close linkage among extrinsic calibration, multi-view fusion and global optimization, which causes fluctuating reconstruction results, cumulative errors and poor system expandability. We propose a decoupled framework for calibration and stateless real-time multi-view point cloud fusion (FUSE-Flow), a framework with two collaborative components: geometry-aligned multi-view extrinsic calibration (GMAC) and reliability-guided multi-view point cloud fusion (FUSE). This split design avoids conflicting optimization objectives for targeted improvement. The GMAC module refines camera extrinsics via geometric constraints and multi-view reconstruction transformers, enabling accurate sparse-view calibration without calibration targets, dense images or global bundle adjustment. The FUSE module integrates confidence weighting and adaptive spatial hashing for stateless fusion, ensuring linear time and memory consumption. The two modules mutually reinforce each other: accurate camera poses boost fusion accuracy, and confidence-aware fusion corrects calibration biases. Validated on public datasets and real camera setups, FUSE-Flow outperforms mainstream real-time reconstruction methods in visual effect, dynamic stability and scalability, providing a practical solution for large-scale real-time 3D reconstruction.
arXiv:2606.04277v1 Announce Type: cross
Abstract: In quantum technologies, continuous-variable systems offer advantages over their discrete counterparts. However, continuous-variable tomography suffers from exponentially growing sample complexity. We propose protocols using quantum mirrors to transfer the complete information of incident photonic states onto a control atomic system. This enables full photonic state characterization through measurements on the control atom alone, realized via kernel functions, direct wavefunction reconstruction, and pointwise Wigner function measurements. Our approach overcomes the limitations of conventional photon counting, statistical inference, and inverse transformation, providing a robust framework for benchmarking and verifying non-Gaussian states in continuous-variable quantum optics.
arXiv:2606.04324v1 Announce Type: new
Abstract: One of the primary challenges in Bayesian inference on the parameters of a diffusion model from discrete observations is the unavailability of an analytical expression for the transition density function between consecutive observation times, which is needed to derive the likelihood function.
Extending previous studies that solve Fokker-Planck (FP) type partial differential equations with Normalizing Flows, we propose a new Normalizing Flow architecture to learn the transition density function of the diffusion process between two observation times. We do so by solving in a Neural Galerkin framework the associated FP equation with a Dirac mass as initial condition, over a specified training distribution of the initial datum and the coefficients of the diffusion. We specifically focus on processes whose diffusion matrix vanishes in certain inaccessible boundary regions, such as Stochastic Volatility models that satisfy a Feller condition.
The product of the obtained transition densities evaluated along the observed trajectory approximates the likelihood function, thereby enabling cheap posterior sampling via Markov chain Monte Carlo (MCMC). After the offline training phase, inference becomes significantly more efficient, as it avoids the need to solve the FP equation in real time for each parameter proposed by the MCMC sampler or to rely on other likelihood-free methods for Bayesian inference that involve repeated simulation of diffusion bridges.
arXiv:2606.04174v1 Announce Type: cross
Abstract: Computed Axial Lithography is a volumetric additive manufacturing method that selectively cures photosensitive resin through the 3D superposition of patterns of light, offering advantages over layer-based processes including rapid print times, reduced layer artifacts, and compatibility with high-viscosity materials. However, diffusive effects, primarily those of free-radical quenchers such as oxygen, blur the boundary between cured and uncured regions, limiting resolution and preventing the reproduction of sharp, high-spatial-frequency features. By comparing micro-CT data to computational dose models convolved with kernels across a range of diffusivities, we establish a framework for extracting a single diffusion kernel from any standard uncorrected print to account for all observed deviations from the target. In this work, we correct diffusion-induced blurring by co-optimizing for its effects alongside the inherent blur of the computed tomography reconstruction, demonstrating improved fidelity over previous approaches of pre-compensating the target geometry via deconvolution.
arXiv:2606.04110v1 Announce Type: new
Abstract: Online evaluation of ranking and retrieval systems often relies on downstream monetization metrics such as app revenue or creator earnings. These metrics are typically heavy-tailed, with a small fraction of users dominating both mean and variance, leading to low statistical power and unreliable conclusions in A/B experiments -- especially under limited traffic.
We present a practical framework for variance reduction in online experiments by combining post-stratification with CUPED. Our approach leverages pre-experiment covariates to improve the sensitivity of monetization experiments without requiring additional traffic. Deployed at ShareChat across ranking-driven monetization experiments, the method substantially reduces variance and improves decision stability, achieving equivalent statistical confidence with ~45\% less traffic than standard metrics. We further discuss practical design choices, guardrails, and limitations, providing guidance on when post-stratification is appropriate for real-world information retrieval and Recommendation systems.
arXiv:2606.04104v1 Announce Type: new
Abstract: Agent systems execute through runtimes with very different control points: local coding tools, framework SDKs, managed agent platforms, API gateways, and observer-only integrations. A high-risk action such as publishing data externally may therefore appear as a shell command in one runtime, a tool call in another, and a hosted session transition in a third. This makes it difficult to answer a basic governance question consistently: what action was authorized, under whose authority, with what approval semantics, and with what evidence after execution?
This paper presents Proof-Carrying Agent Actions (PCAA), a runtime-neutral governance model centered on an action certificate rather than on a vendor-native session record. PCAA organizes control around five checkpoints: pre-action admissibility, action open, assumption capture, approval, and outcome closure. It binds these checkpoints to a portable action envelope, runtime and approval receipts, and replay-ready proof. The model is extended in two practical ways: the certificate is externality-aware, carrying boundary facts such as destination visibility and account provenance, and approval is described by explicit enforceability classes rather than by a single reviewed or unreviewed bit.
We study the model through a reference implementation in a heterogeneous agent control plane and a disclosure-bounded evaluation protocol. On a protected benchmark expanded from 24 executable seeds to 96 traces across four runtime families, PCAA preserves route quality while exposing distinct failure modes under ablation. The paper contributes a systems formulation of runtime governance around certificate-bearing actions and an implementation-grounded account of how that formulation can remain portable under runtime churn without collapsing into vendor-specific control surfaces.
arXiv:2606.04173v1 Announce Type: cross
Abstract: It is well-known that any class of simple graphs, that is characterized by finitely many forbidden minors, also admits a characterization by finitely many forbidden topological minors; furthermore, the list of forbidden topological minors may be derived from the list of forbidden minors.
We prove a similar result in Matching Theory. Our Main Theorem states that any class of matching covered graphs, that is characterized by finitely many forbidden $S$-minors that are cubic, also admits a characterization by finitely many forbidden conformal minors that are cubic as well; once again, the list of forbidden conformal minors may be derived from the list of forbidden $S$-minors.
In order to establish the above, we first prove that every matching covered graph has one of two graphs as a conformal minor -- either $K_4$, or the $\Theta$ graph (that is, two vertices joined by three edges). (In fact, we need and prove a much stronger statement.) This is reminiscent of a theorem due to Lov\'asz: every nonbipartite matching covered graph has one of two graphs as a conformal minor -- either $K_4$, or the triangular prism $\overline{C_6}$.
As applications of our Main Theorem, we deduce known 'forbidden conformal minor characterizations' of pfaffian near-bipartite graphs, and of pfaffian solid graphs, using their respective known 'forbidden $S$-minor characterizations'.
arXiv:2606.04138v1 Announce Type: cross
Abstract: The coordinated interactions between organisms enhance collective functionality, a feature that artificial systems such as enzymatic nanomotors seek to replicate. A key objective, yet still a major challenge, is to achieve chemical communication among nanomotors. Progress has been limited by the difficulties in verifying effective signaling processes, including chemical signal propagation and the response of receiving nanomotors. Here, we address this challenge using an enzymatic nanomotor system that demonstrates communication between two populations through generically non-reciprocal phoretic response. A primary swarm of glucose-responsive nanomotors migrates toward a glucose gradient while producing H2O2 as a diffusible communication signal. This self-generated chemical gradient then acts as a chemoattractant for a secondary swarm of catalase-powered nanomotors. Through carefully designed experiments, we visualize the propagating H2O2 gradient and quantify the spatiotemporal response of the receiver nanomotors to the chemical front. Combined experimental and theoretical analysis has revealed that the synergy between different combinations of chemo-attractive and chemo-repulsive mobilities and catalytic rates of consumption and production of substrates and products gives rise to a wealth of different collective responses in the system. This work represents a step toward programmable synthetic systems at the collective level, broadening the functionality of chemical nanomotors and opening opportunities for future hybrid living-synthetic systems.
arXiv:2605.19294v2 Announce Type: replace
Abstract: Vision-Language-Action (VLA) policies increasingly rely on asynchronous inference to hide large-model latency behind ongoing robot motion. While this avoids the stop-and-go behavior of synchronous action-chunk execution, it creates a prediction-execution mismatch: the next chunk is computed from a stale observation at inference start but executed only after the robot and scene have evolved. As a result, actions that fit the prediction-time state can become misaligned with the execution-time state. Existing runtime repair, behavior-cloning, and preference-alignment approaches do not directly teach the policy to resolve this stale-input mismatch. We propose DEFLECT, an offline post-training framework for delay-robust asynchronous VLAs. DEFLECT converts latency-induced mismatch into counterfactual preference supervision: a frozen reference VLA generates a preferred chunk from the future execution-time observation and a rejected chunk from the stale prediction-time observation. The trainable policy scores both chunks under the same deployment-time input, learning to favor execution-time-aligned actions while a supervised fine-tuning anchor preserves the expert action manifold. DEFLECT requires no human preference labels, reward models, online robot rollouts, architectural changes, or additional inference-time computation. Across Kinetix, LIBERO, and three real-robot tasks, DEFLECT improves delay robustness over strong asynchronous VLA baselines, raising high-latency success by up to 6.4 percentage points and achieving a 4.6 percentage-point gain at the longest delay on a real-scale VLA.
arXiv:2606.04267v1 Announce Type: cross
Abstract: The estimation of concentration parameter in Fisher-von Mises-Langevin distribution is the directional statistics analogue of the estimation of the precision matrix for the Gaussian distribution. In this work we show that unbiased estimation of this parameter is impossible. With this realization in hand, we provide an alternative parameterization of the Fisher-von Mises-Langevin distribution in terms of the squared concentration, which we term the intensity. We fruther show that unbiased estimation of thereof is possible, and provide (almost) unbiased estimators thereof in terms of a partial sum U-statistic. We showcase our new estimator on synthetic data, New York taxi trip data, and on spherical word embeddings.
arXiv:2606.05030v1 Announce Type: new
Abstract: Autoregressive chain-of-thought (CoT) reasoning in large language models (LLMs) is fundamentally forward-directed: each step conditions only on prior tokens. This unidirectional inductive bias renders even capable models susceptible to error snowballing, wherein a single logical or arithmetic mistake in an early step irreversibly corrupts the entire reasoning chain. We introduce Teleological Reasoning Infilling (\TRI{}), a training framework that endows decoder-only transformers with a native \emph{goal-conditioned bridging} capability. The key insight is to reframe erroneous reasoning segments as fill-in-the-middle (FIM) tasks: given a verified prefix premise $P$, a verified downstream milestone $S$, and the original query $Q$, the model must synthesise the logical bridge $M$ that connects $P$ to $S$ rigorously and completely. To achieve this with standard causal architectures, we introduce a Prefix-Suffix-Middle (PSM) sequence rearrangement with three non-overlapping sentinel tokens, enabling $M$ to attend to both $P$ and $S$ without any structural modification to the self-attention mechanism. Training proceeds in two stages: (i) Supervised Fine-Tuning (SFT) on symbolically verified $(P, S, M)$ triples extracted from formal mathematics corpora, and (ii) Direct Preference Optimisation (DPO) with a deterministic symbolic verifier (Lean 4 / Python) as the sole reward oracle, eliminating LLM-judge sycophancy. At inference, TRI operates as a surgical repair module within a dual-system loop: a causal draft model generates an initial trace, the verifier pinpoints failures, and TRI infills only the damaged segment, leaving verified sections intact. Comprehensive experiments on three benchmarks demonstrate that TRI achieves state-of-the-art performance across all tasks, while reducing per-problem token expenditure by 31.2%.
arXiv:2606.04123v1 Announce Type: cross
Abstract: Trajectory optimization is a critical component for enabling safe and reliable autonomous operations in space exploration. As space missions increase in frequency, complexity, and scope, there is a growing need to rapidly formulate mathematically sound trajectory optimization problems that accurately reflect mission objectives and operational constraints. However, translating mission intent into tractable analytical formulations for trajectory optimization requires substantial domain expertise. This paper presents a framework that leverages large language models (LLMs) to translate natural language descriptions of mission requirements and constraints into executable trajectory optimization code and corresponding mathematical formulations. Experiments in spacecraft rendezvous scenarios demonstrate a high success rate in reconditioning a convex trajectory optimization problem from semantic mission requirements. Ultimately, this work highlights the potential of LLMs to bridge high-level intent and formal optimization models, enabling more flexible and efficient trajectory design of spacecraft.
arXiv:2606.04093v1 Announce Type: cross
Abstract: In numerical experiments simulating Turbulent Radiative Mixing Layers (TRMLs) it is observed that as the cooling time in the mixed gas, $t_{\rm cool}$, becomes very short compared to the dynamical time of the turbulence, $t_{\rm eddy}/t_{\rm cool} \gg 1$, there is a change in the scaling behavior of the total energy radiated in the TRML as a function of this ratio, also known as the Damk\"{o}hler number, ${\rm Da} \equiv t_{\rm eddy}/t_{\rm cool}$, from $\dot{E}_{\rm cool} \propto {\rm Da}^{1/2}$ to $\dot{E}_{\rm cool} \propto {\rm Da}^{1/4}$. The latter, so-called "fast-cooling," regime is of particular interest as many astrophysical mixing layers lie in this regime. We demonstrate that the origin of this change is the suppression of turbulent folding of the surface by the ram-pressure of the inflowing gas, which becomes much greater than the turbulent pressure in this regime. We present an argument that reproduces the $\dot{E}_{\rm cool} \propto {\rm Da}^{1/4}$ behavior by appealing to the suppression of the fractal structure of the interface by the ram-pressure of the inflowing gas.