arXiv:2606.17288v2 Announce Type: replace
Abstract: Perfect resource placements in dense Eisenstein--Jacobi (EJ) networks partition the network into hexagonal radius-$t$ service cells. This paper studies local repair of such placements after resource failures. For one failed resource, we prove that one replacement cannot cover the failed hexagon and two always suffice, giving $\rho_{\mathrm{EJ}}(t)=2$ for all $t\ge1$. Among minimum-size repairs, the sharp minimum-overlap formula $\Omega_{\mathrm{EJ}}(t)=t^2$ follows from the three-strip geometry of EJ balls. For two failed resources, independent repair gives a four-replacement upper bound, but unlike the Gaussian case EJ repair is not always additive: two infinite neighboring displacement families admit three-replacement repairs, proved optimal by a two-ball impossibility argument. Additive behavior is established algebraically via endpoint-rigidity and diagonal-corridor theorems. For $q$ failed resources, independent canonical repair gives a universal $2q$ upper bound, exact when failed cells are pairwise more than $4t$ apart. Dense cluster subadditivity is proved for infinite four-fault and six-fault families with exact repair numbers four and five, giving savings of four and seven over independent repair. An exact inclusion--exclusion identity governs repeated coverage for arbitrary multi-fault repairs. An audit over 19,400 instances confirms widespread subadditivity. EJ local repair is structurally distinct from the Gaussian case: the one-fault overlap is quadratic, two-fault repair can be non-additive, and clustered repairs reuse replacement balls across multiple failed cells.
Science Journals
arXiv:2606.17386v2 Announce Type: replace
Abstract: End-to-end autonomous driving has achieved state-of-the-art performance on benchmarks and real-world deployments. Its standard training recipe, however, is expensive across all stages: collecting and labeling millions of driving frames is costly, and closed-loop RL on images is bottlenecked by the per-step cost of photorealistic rendering plus a forward pass through a large vision backbone. Self-play in vectorized simulators changes the economics: millions of rollout steps per second, and a state distribution naturally rich in collisions, near-misses, and recoveries that no driving log contains. Our approach exploits this asymmetry by decoupling learning to drive from learning to see. We pretrain a single policy by self-play, then align its latent space with a pretrained vision backbone, through the action KL divergence and a batch-relational low-rank structural loss. The action target comes from the self-play policy, so alignment never supervises against a logged trajectory: a paired dataset of (image, scene-state) frames suffices, with no need for the curated expert demonstrations that imitation pretraining is built on. On photorealistic 3D Gaussian splatting closed-loop scenarios, the resulting end-to-end policy matches or exceeds prior end-to-end methods.
arXiv:2606.17930v3 Announce Type: replace
Abstract: AI evaluations are shifting toward harder tasks that benefit from longer trajectories involving tool use and iterative problem solving. As a result, performance is increasingly sensitive to the amount and allocation of compute available at test time ("inference compute"). Yet many evaluations still report performance at a single restrictive budget, meaning that low scores may reflect the evaluation setup rather than the model's underlying capability. To test this, we evaluate up to 12 frontier language models on seven challenging benchmarks spanning software engineering, mathematics, medicine, and cybersecurity. We use a controlled setup combining three simple inference-scaling interventions: larger token budgets, context compaction, and repeated submission attempts, guided either by the model itself or by minimal correctness feedback. We find three main results. First, larger token budgets substantially improve performance on benchmarks across multiple domains, including cybersecurity, FrontierMath, Humanity's Last Exam, and TerminalBench. Second, fixed-budget evaluations can increasingly understate frontier capability as models advance. Newer models reach higher performance at large budgets, where they unlock harder tasks and solve them more reliably. Third, benchmarks differ in which inference-scaling methods help most: repeated submission broadly improves performance, but the value of larger token budgets, external feedback, and parallel attempts varies by benchmark. Overall, our results show that benchmark scores are protocol-dependent. We therefore argue that evaluations should report capability as a function of inference-time compute, specify protocol choices explicitly, and compare model generations over a large shared compute range at matched budgets, especially in safety- or policy-relevant settings.
arXiv:2606.18918v4 Announce Type: replace
Abstract: This paper investigates the computational complexity of verification problems for Binarized Neural Networks (BNNs), in which activations and weights are binary. Specifically, we study three verification problems. First, we prove that checking the satisfiability of a linear property for a BNN is NP-complete via a reduction from the Boolean Satisfiability (SAT) problem. Second, we show that verifying robustness under non-uniform image occlusion is NP-complete through a reduction from SAT. Finally, we demonstrate that uniform occlusion induces a piecewise-constant structure in the network output, which enables the design of a polynomial-time algorithm for robustness verification.
arXiv:2606.19729v2 Announce Type: replace
Abstract: Planning under uncertainty is an essential capability for autonomous robots. The Partially Observable Markov Decision Process (POMDP) provides a powerful framework for such a capability. Although POMDP-based planning has advanced significantly, its application to real-world problems is often limited by the difficulty of obtaining faithful POMDP models. We present Vectorized Online planning wIth Learned diffusion model for POMDP Agents (VOiLA), a framework that learns task-agnostic POMDP models for online planning under uncertainty. VOiLA learns transition and observation samplers using conditional diffusion models and learns observation-likelihood models for particle-based belief updates. To enable efficient online planning, the diffusion samplers are distilled into compact feedforward generators and integrated with Vectorized Online POMDP Planner (VOPP), an online POMDP planner designed to leverage GPU parallelization. Experimental results indicate the distillation strategy reduces sampling cost by up to nearly three orders of magnitude, making learned generative POMDP models practical for online planning. Evaluation of VOiLA on three benchmark problems indicate that VOiLA achieves equal or better performance than Recurrent Soft Actor Critic while using less than 10% training data, and generalizes much better to unseen environment configurations. Physical robot evaluation indicates VOiLA uses the models learned using only simulated data and generates a policy that successfully accomplish the task in 10 of 10 runs.
arXiv:2606.19817v2 Announce Type: replace
Abstract: Synthetic images are increasingly used to augment scarce real data for object detection. However, not all synthetic sets help equally, and the only way to know a set's value is to train a detector on it, which is slow and demands dense annotation. We ask whether a training-free metric can instead rank candidate synthetic training sets by their downstream utility. Existing image-set metrics such as FID, KID, and MMD compare two feature distributions with a single global statistic, which we show is mis-specified for detection-data selection in two ways: it is blind to per-image composition (object count, box scale, class mix), and even at fixed composition its global averaging washes out the appearance differences that separate high-mAP pools from low-mAP ones. We propose Conditional-Composition Domain Match (CCDM), which converts any feature-space distance into a composition-stratified comparison, matching candidate and target within metadata-defined strata without training a detector. On COCO and VisDrone-DET, the best CCDM variant ranks 19 candidate training sets in strong agreement with YOLOv8 mAP (Spearman \r{ho} = 0.97 and 0.96), outperforming FID, KID, and MMD. Furthermore, CCDM holds when reference metadata comes from detector pseudo-labels rather than ground-truth boxes.
arXiv:2606.21005v2 Announce Type: replace
Abstract: Scientific discovery workflows often depend on structured curation from the literature. This is difficult for current agents because the key evidence is scattered across long text, dense tables, and figures, and the final records often require reasoning across multiple evidence fragments rather than copying a single span. We study scientific curation from multimodal sources and introduce Beaver, an agent harness that extracts structured information from scientific papers while preserving provenance to the supporting evidence. Beaver combines a frontier agent with multimodal evidence tooling, task scaffolding, and artifact-grounded autoresearch. These components turn curation into a staged, auditable workflow and enable an iterative evaluate--diagnose--revise loop, where persistent run artifacts expose stage-localized failures and guide harness updates. Experiments show that Beaver reaches 81.0 on Gold-Referenced Attribute Score (GRAS), an attribute-level measure of agreement with gold curated records, outperforming frontier agents by over 23 absolute points. Ablations show that task scaffolding, multimodal evidence tooling, and provenance traces each contribute meaningfully to performance, while attribute-level analysis shows the largest gains on high-value attributes that require cross-modal reasoning and normalization. These results show that, for scientific curation from papers with multimodal evidence, harness design is a central determinant of agent performance.
arXiv:2606.21317v2 Announce Type: replace
Abstract: Recent work has raised concerns about the influence of sycophantic AI on user judgment and relationships. One proposed mitigation, which has received regulatory attention, is to warn users about potentially harmful AI behaviors such as sycophancy. In a preregistered experiment in which participants (N = 2,610) discussed real interpersonal conflicts with an AI system, we test whether warning labels mitigate sycophancy's influence. We find that a basic AI disclosure (``This chatbot is AI'') has no detectable effect. Labeling the system as sycophantic (``...may agree with you and validate you even when you are wrong...'') does shift users' perceptions, reducing perceived objectivity and trust, but it does not reliably reduce sycophancy's influence on users' self-perceived rightness or their willingness to repair the conflict. Our results reveal a gap between AI perception and AI influence: by shifting perception without reducing influence, warning-based interventions may offer a false sense of protection. Addressing the harms of sycophancy will therefore require understanding the specific mechanisms through which it shapes judgment, and improving model behavior itself.
arXiv:2606.22030v2 Announce Type: replace
Abstract: We investigate when belief-based memory actually improves large language model (LLM) agents. Our vehicle is Nous, a long-term memory architecture that represents each entity-attribute pair as a categorical probability distribution updated through closed-form Bayesian inference, with information-theoretic surprise driving belief revision and entropy-based forgetting. A controlled ablation on the LoCoMo benchmark shows that Bayesian belief updating alone provides little benefit over naive last-write-wins because existing conversational memory benchmarks rarely contain contradictory or differently reliable evidence. We then introduce reliability-conditioned updating, estimating per-observation reliability from epistemic language, and show on a controlled contradiction benchmark that belief updating substantially outperforms last-write-wins and raw-memory retrieval when observations differ in trustworthiness. Because content-derived reliability is itself vulnerable to manipulation, we further propose provenance-capped belief updating, where trust is bounded by source provenance rather than textual confidence. Under controlled memory-poisoning experiments, this approach resists volumetric poisoning attacks while revealing the utility costs and implementation requirements of provenance-aware memory. Finally, we quantify a 27.5-point discrepancy between strict token-F1 and LLM-as-judge evaluation on identical outputs, highlighting important reproducibility concerns for long-term memory benchmarks. Our results suggest that probabilistic belief-based memory is most beneficial in environments requiring reasoning over conflicting and differently trustworthy evidence, rather than conventional conversational recall alone.
arXiv:2606.23040v2 Announce Type: replace
Abstract: Inductive proofs can be represented by proof schemata, a formalism that represents infinite sequences of proofs by recursive definitions. Since proof schemata avoid the explicit application of induction rules, they admit novel applications, one of which is the realization of Herbrand's theorem in the presence of induction. In this paper, we develop a new type of proof schema based on point transition systems. For skolemized proof schemata without quantified cuts, so-called Herbrand systems, that is, schemata of Herbrand instances of quantified formulas, can be computed. Herbrand systems also allow the representation of schemata of Herbrand sequents, thereby realizing Herbrand's theorem for proof schemata. We compare proof schemata with cyclic proofs and define a transformation from a large class of cyclic proofs to proof schemata. Finally, we show that proof schemata based on point transition systems are capable of proving the 2-Hydra statement, a well-known example that is provable by the cyclic proof system CLKID\omega but not in LKID.
arXiv:2606.25286v3 Announce Type: replace
Abstract: When institutions decide by consensus, the official record shows agreement but hides who shaped what was decided. We introduce a way to recover that hidden structure from the one trace consensus cannot suppress: the documentary record of what actors choose to work on. Adapting tools from economic complexity, we map a ``space of concerns'' in which issues lie close when the same actors repeatedly specialize in both -- turning a flat agenda into a measurable topology of attention. Across six decades of the Antarctic Treaty (6,591 documents, 66 actors), engagement is structured, local, and persistent, and the most specialized actors produce binding law at roughly five times the baseline rate. The approach generalizes to any document-rich consensus forum, showing that unanimity does not erase political structure -- it relocates it upstream, into the organization of attention.
arXiv:2606.27238v2 Announce Type: replace
Abstract: Production of useful chemicals using photoelectrochemical biohybrid devices offers an environmentally friendly alternative to existing energetically demanding processes. These devices exploit light-driven charge separation, e.g. by a photosystem, and require efficient electron transfer to a tailored redox enzyme cascade. Here we demonstrate that electron transfer efficiency can be increased by confining the photosystem with the redox protein inside a self-assembling, virus-based nanocontainer. The photosynthetic system from the phototrophic bacterium Cereibacter sphaeroides and cytochrome c were conjugated to a bacteriophage P22 scaffolding protein and co-incorporated into the 50 nm diameter virus shell in vitro. The porous shell confined the macromolecular components for efficient electron transfer while allowing free exchange of small electron mediators. Sustainable and accelerated light-driven electron transfer between the encapsulated components was confirmed by optical spectroscopy. This self-assembly system presents a versatile platform for developing nanoreactors that combine photosystems with complex redox pathways.
arXiv:2606.28149v2 Announce Type: replace
Abstract: In-context segmentation (ICS) requires a model to segment target regions in a query image using only a few reference images and their corresponding masks, without updating any parameters. Despite recent progress, prior ICS studies have largely overlooked a critical aspect: system robustness, ie, whether the model can produce stable segmentation results for the same query under different references. In this work, we revisit ICS from the robustness perspective and introduce a novel paradigm, Concept-Guided In-Context Segmentation (CG-ICS), which performs segmentation by extracting high-level semantic concepts from references rather than relying solely on low-level visual matching. Specifically, CG-ICS introduces a concept reasoning module that uses an MLLM to propose candidates and a SAM3-driven scoring function with tree-search refinement to select reliable textual concepts, together with a parallel visual exemplar route that provides query-side spatial grounding via a simple context construction. Both the textual concept and the visual exemplar are then used to activate the segmentation capability of a frozen SAM3 backbone. Extensive experiments on standard ICS benchmarks demonstrate that CG-ICS not only achieves state-of-the-art accuracy but also substantially improves robustness, yielding a more reliable ICS system with significantly reduced variance across diverse reference choices. Code is available at https://github.com/Kakarot1103/CG-ICS.
arXiv:2606.29059v2 Announce Type: replace
Abstract: World modeling requires forecasting uncertain futures while preserving information useful for
downstream perception. Existing visual world models often struggle to satisfy both goals:
VAE-based stochastic models operate in low-dimensional reconstruction latents, which can
limit perception performance, while deterministic predictors using strong pretrained features
collapse multimodal futures into a single blurry mean. In this work, we propose FlowWM, a
stochastic world model that performs flow matching directly within pretrained feature space
(e.g., DINOv3). This is challenging because pretrained features are substantially
high-dimensional, making standard diffusion recipes suboptimal. To address this, we
investigate the design choices needed for feature-space flow matching and introduce a
differentiable one-step projection mechanism that enables efficient training with temporal
consistency and task-driven objectives. We evaluate FlowWM on two benchmarks: a synthetic
benchmark for systematic evaluation of accuracy and diversity, and a real-world benchmark
FuturePerception. FlowWM improves perception performance, mode coverage, and horizon
robustness, validating our proposed design for stochastic world modeling in high-dimensional
feature spaces.
arXiv:2606.29538v3 Announce Type: replace
Abstract: Skills are a useful abstraction for software agents, turning human and agent experience into reusable procedural knowledge. Yet existing skill libraries are mostly hand-written, text-centric, or derived from agent traces, leaving tutorial videos and other multimodal human resources largely underused. We present RESOURCE2SKILL, a framework that distills multimodal resources, including tutorial videos, repositories, articles, and reference artifacts, into executable skills for software agents. RESOURCE2SKILL organizes these skills as a hierarchical multimodal Skill Wiki, where each entry combines structured text, code, visual examples, metadata, and provenance. This design preserves complementary signals from different resources: videos capture temporal operations and visual effects, code captures executable tool patterns, and articles or artifacts provide conceptual and stylistic grounding. At inference time, agents retrieve and compose relevant skills from the wiki; when coverage is insufficient, the same construction operator can acquire new skills online. Across seven practical authoring domains, RESOURCE2SKILL improves average overall score by +11.9 percentage points over no-skill agents and outperforms strong harness baselines in 26 of 28 main-aggregate model-domain cells. Ablations confirm the value of multimodal skill format, hierarchical organization, source diversity, selection strategy, and online acquisition.
arXiv:2606.29814v2 Announce Type: replace
Abstract: We propose Nemotron-Labs-Diffusion-Image, a state-of-the-art masked discrete diffusion model (MDM) for high-resolution text-to-image synthesis. Compared with prior work on masked image generation, Nemotron-Labs-Diffusion-Image addresses two key challenges. First, unlike continuous diffusion models which progressively refine latent representations across the entire image, standard MDMs lack self-correcting capability because discrete tokens cannot be modified once they are unmasked. Second, although increasing the vocabulary size of discrete image tokenizers improves reconstruction fidelity, it introduces optimization difficulties for generative modeling as the per-token training signal becomes increasingly sparse. To address the first challenge, Nemotron-Labs-Diffusion-Image incorporates a token-editing mechanism that enables the model to dynamically revise already-unmasked tokens during inference, similar to how a sculptor iteratively refines their work. To tackle the second challenge, we propose a Grouped Cross-Entropy (GCE) objective that assigns positive learning signals to tokens neighboring the ground truth in embedding space, thereby alleviating signal sparsity. To further improve training efficiency, we implement a custom fused operator for GCE that significantly reduces VRAM usage in large-vocabulary settings. Experimental results demonstrate that these innovations substantially improve both training efficiency and image fidelity of masked discrete image generators, achieving a score of 0.90 on GenEval, 86.9 on DPG and 10.76 of HPSv3.
arXiv:2606.29932v3 Announce Type: replace
Abstract: Long-horizon strategic planning in complex strategy games requires coordinating tightly coupled decision domains, including technology, economy, diplomacy, and military, across hundreds of turns under imperfect information. Existing LLM-based agents face three challenges in this setting. They recover little relational structure, such as distance or threat, from raw coordinate observations. They serialize the entire growing game state into every prompt and plan all domains in a single output, so the context eventually overflows and urgent domains crowd out long-term ones. The only reward is a lagged final score, which provides no progress signal within or across games.<br/><br/>We present SAGA, an LLM multi-agent framework that addresses these challenges with three mechanisms: a Map-Semantic Scene Graph that renders typed entity relations as concise per-entity text, a Tool-Augmented Planner that retrieves state on demand and routes disjoint per-domain directives to specialist controllers, and a Dual-Horizon Feedback Loop that sets intermediate goals within a game and distills causal lessons across games. On CivRealm, a benchmark built on the strategy game FreeCiv, SAGA attains the highest mean final score among six methods, the statistically strongest gains on infrastructure, and a 27% reduction in output tokens. With cross-game evolution enabled, it reaches the highest score over five successive games.
arXiv:2606.30005v3 Announce Type: replace
Abstract: Long-horizon tool agents are bottlenecked by how their context grows toward the limits of the context window. Recent systems make context management agent- or system-controlled, but they either learn a compression policy that discards evidence or manage context in a layer the agent never sees. We argue both leave a more basic gap unaddressed. Frontier language models are proprioceptively blind to their own context. From the prompt alone they cannot see how large, how old, or how used each block is, the signals a keep-or-drop decision needs. We hypothesize that competent context management is already latent in capable models, and that what is missing is not a learned policy but an interface exposing this state. We introduce VISTA (Visible Internal State for Tool Agents), a training-free, model-agnostic layer that represents working memory as typed, addressable blocks, surfaces a runtime dashboard of per-block token usage, recency, and access history, and archives blocks as recoverable full-fidelity payloads. On LOCA-Bench, BrowseComp-Plus, and GAIA, the same untrained interface transfers across 1M-, 100K-, and 10K-scale trajectories. On LOCA-Bench it improves four backbones and lifts Gemini-3-Flash from 22.7 to 50.7%. The lift grows with context pressure and transfers across backbones. Ablations further confirm that the dashboard matters beyond archive and recovery tools.
arXiv:2606.30252v2 Announce Type: replace
Abstract: Inoculation prompting is a selective-generalization technique used against Emergent Misalignment. We introduce inoculation adapters (IA), a family of methods that similarly reduce the optimization pressure to learn undesired traits by strengthening those traits during training. Inoculation adapters are LoRAs that are trained and used in three steps: (1) trained on undesired traits; (2) attached frozen while a separate task adapter is trained on data exhibiting both desired and undesired traits; (3) the IA is discarded at deployment, while only the task adapter is kept. We compare inoculation adapters with four selective-generalization baselines: inoculation prompting, preventative steering, Concept Ablation Fine-Tuning (CAFT), and KL regularization. Across nine setups and five model families, the inoculation adapter family spans a new Pareto frontier of desired trait retention vs. undesired trait suppression, although given wide confidence intervals the magnitude of improvement remains uncertain. Inoculation adapters also avoid two drawbacks of inoculation prompting: they can suppress capabilities and traits that cannot be reliably elicited by a prompt, and they introduce fewer surprising backdoors. However, no IA variant optimizes all objectives perfectly; gains in desired-trait generalization are generally accompanied by weaker suppression of the undesired trait and increased backdoor occurrence.
arXiv:2606.31456v2 Announce Type: replace
Abstract: With an increasing number of Object Detection (OD) models being deployed on edge devices, Zero-Shot Quantization for OD (ZSQ-OD) aims to quantize these models when access to the original training data is prohibited. Existing research on Zero-Shot Quantization-Aware Training (QAT) for OD synthesizes training sets through noise optimization. However, this approach struggles to maintain performance in low-bit regions. In this paper, we introduce GoodQ (Generative off-the-shelf models for object detector Quantization), a QAT pipeline that utilizes off-the-shelf generative models to construct a training set. We first identify three challenges that arise when introducing a generative model to the ZSQ-OD task: 1) each image contains dense information with multiple instances, 2) the class-wise distribution in the original dataset is imbalanced, and 3) the pseudo-labels assigned to the generated images can potentially act as noisy signals during QAT. GoodQ addresses these challenges by 1) introducing an Information-Dense Prompting strategy to generate multi-instance images, 2) applying Intrinsic Distribution-Aware Selection to match the pretrained class distribution, and 3) employing Teacher-guided Adaptive Noise Reduction to mitigate noise arising from the QAT process. Our framework achieves state-of-the-art performance in low-bit ZSQ (W4A4) and extends quantization to extreme bit-widths (W3A3). Furthermore, we conduct an extensive analysis to uncover the underlying factors contributing to the efficacy of GoodQ.
arXiv:2606.31693v2 Announce Type: replace
Abstract: The wave of AI-native applications is moving shopping beyond page- and feed-based browsing toward intent-driven experiences orchestrated by LLM agents. A common design wraps an LLM around existing search and recommendation pipelines, forcing complex intents through low-bandwidth retrieval or ranking interfaces and leaving a gap between language understanding and item-space fulfillment. Generative recommendation gives LLMs a direct item-space interface through semantic IDs (SIDs), but existing models mainly generate candidates for retrieval rather than translate flexible intents into item-space outcomes. We propose ShopX to address this bottleneck by unifying intent understanding, execution planning, and flexible SID-native item-space operations into a single foundation model. We deploy ShopX in agentic shopping workflows through a model-native item-fulfillment framework with a serving harness that defines a model-facing action protocol and exposes support surfaces for context access, catalog grounding, and state management. Within this framework, ShopX plans and composes SID-based item-space operations such as SID beam-search retrieval, listwise ranking, or product bundling. This model-centric design reduces lossy hand-offs between agent orchestration and item-space execution. To build ShopX, we design semantically recoverable, LLM-operable SIDs and a training recipe that equips a general LLM for flexible multi-turn item-space fulfillment while retaining the knowledge and instruction-following abilities needed by a shopping agent. We evaluate the ShopX framework against tool-mediated agentic systems on single- and multi-turn fulfillment tasks derived from anonymized Taobao production logs, showing that model-native fulfillment improves overall framework behavior, especially on complex or ambiguous requests.
arXiv:2607.00724v3 Announce Type: replace
Abstract: Multilingual fluency often invites a stronger assumption: a model that can speak a user's language must also understand the culture encoded by that language. We call this the Illusion of Cultural Alignment. To test this assumption directly, we introduce MSQA, a benchmark of 1,064 natively sourced questions across 11 language groups, five cultural dimensions, and three difficulty tiers. Unlike translated benchmarks, MSQA targets locally grounded knowledge and reduces shortcuts from English-centric cross-lingual transfer. Evaluating 18 LLMs, we find substantial cultural degradation and a pronounced Locality Effect: cultural competence tracks pre-training exposure more closely than general reasoning ability. We further show that common inference-time remedies do not dissolve the illusion. Models remain overconfident on unfamiliar cultural questions, repeated sampling yields unstable rather than reliable correctness, and retrieval augmentation helps unevenly on long-tail facts. These findings indicate that cultural alignment cannot be inferred from multilingual ability alone and requires deeper intervention than calibration, sampling, or retrieval at inference time
arXiv:2607.01106v2 Announce Type: replace
Abstract: Distributed trajectory estimation arises in many applications across robotics, but existing implementations typically do not consider asynchrony in agents' communications and computations. Therefore, we propose an asynchronous block coordinate descent algorithm for distributed trajectory estimation. We consider a team of agents that observes a team of robots and estimates the robots' states over a sliding window. The agents solve an approximation of the maximum a posteriori estimation problem, which we derive. We show this approximation introduces negligible errors and eliminates up to 96.9% of communications among agents. Next, we prove that agents' iterates converge exponentially fast to the optimal estimate of the robots' states. Simulations show that this approach has up to 64% less error than a comparable state-of-the-art algorithm. Experiments on mobile robots show this approach is robust to delays whose lengths span three orders of magnitude.
arXiv:2607.01153v2 Announce Type: replace
Abstract: Safety evaluations for language models increasingly depend on judgments about ambiguous natural-language behaviour: whether a model has followed an instruction, refused appropriately, complied with a policy, resisted an embedded command, or misreported progress in an agentic task. Existing benchmarks often compress these distinctions into pass/fail labels, obscuring whether failures arise from capability limits, policy ambiguity, instruction conflict, scaffold failure, or unstable evaluator judgments.
This paper introduces adversarial pragmatics as a benchmark and annotation protocol for evaluating model behaviour under instruction conflict, embedded commands, quotation, scope ambiguity, deixis, indirect speech acts, and multi-turn agent transcripts.
The contribution is empirical and methodological: a linguistically controlled taxonomy, an 18-item seed benchmark with validator-enforced metadata, a 54-row local seed pilot, an expert-evaluation protocol distinguishing task success, policy compliance, safety risk, refusal outcome, and evaluator confidence, and metrics for judge validity, diagnostic ambiguity, and taxonomy drift. The benchmark treats labels as inference licenses: it tests whether safety-relevant categories project across paraphrase, wrapper, model, and judge condition. In the pilot, a rubric-aided LLM judge graded its own outputs with expected-behaviour fields visible and still missed the safety-relevant minority classes.
arXiv:2607.01211v2 Announce Type: replace
Abstract: Repository-level performance-optimization benchmarks such as GSO, SWE-Perf and SWE-fficiency evaluate coding agents by applying patches to real repositories and comparing runtime against unoptimized baselines and official reference patches. Their leaderboard scores are increasingly used as evidence of coding-agent progress, but those scores can conflate runtime instability, benchmark-specific scoring rules, and how many tasks are already solved by at least one public submission. We audit these issues across the three benchmarks. First, we replay the official reference patches for 740 code optimization tasks across four common types of Google Cloud machines. Most benchmark tasks can be replayed, but their reference patches satisfy the original benchmark validity rules in every cross-machine replay for only 39/102 GSO tasks, 11/140 SWE-Perf tasks, and 411/498 SWE-fficiency tasks; SWE-Perf is especially fragile because many reference patches produce close-to-zero runtime changes. Second, we show that public submission rankings depend strongly on the benchmark scoring rule. Among eight public submissions shared by GSO and SWE-fficiency, the official rankings disagree on 9 of 28 pairwise submission comparisons, and SWE-fficiency's leaderboard scoring rule assigns the worst ten tasks overly high score weights of 58.5%-82.8%. Third, looking across 10 public submissions for each task, we find that at least one submission matches or beats the reference patch on 85.3% (384/450) of replay-valid GSO and SWE-fficiency tasks, and beats the unoptimized base code on 99.8% (449/450). Our study complements leaderboard scores by identifying tasks with more reliable performance signals, quantifying per-task score contributions, and exposing the remaining performance gaps that are hidden by aggregate rankings.