PyTorchKR Papers

최신 논문

총 568,830편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2611.00580v2cs.AI

Scaling Science: Toward a Chain Reaction of Discovery

Scientific discovery advances when evidence from different investigations is brought into contact. Yet paper-centric communication leaves experimental results fragmented, delayed, or unreported, creating a "scientific memory wall" between growing computational capabilities and the evidence accessible to AI systems. Breaking this wall requires a paradigm shift in AI for Science: moving from passively mining questions out of existing datasets to actively generating and recombining experimental data around unresolved scientific questions. In a network of intelligent laboratories, a failed or overlooked experiment in one domain could trigger a hypothesis in another, whose testing generates fresh evidence that stimulates further investigations. To enable these evidence-driven chain reactions, we propose Evolving Scientific Intelligence (ESI), an architectural blueprint that connects shared scientific memory with distributed experimental loops. ESI preserves experimental records -- including negative and anomalous results -- in computable representations, allowing models to recombine evidence across domains into testable hypotheses. Rather than following fixed screening pipelines, the system jointly selects candidates and evaluation methods according to scientific value, reliability, and cost, directing distributed laboratories to produce the exact evidence needed next. Their physical feedback revises both hypotheses and evaluators, while slower meta-control reassesses research objectives and decision criteria to guard against self-reinforcing errors and misguided optimization. We call for an Information Manhattan Project to build this infrastructure for scaling science through experimentally grounded chains of discovery.

arXiv:2611.00403v2cs.LG

MXCode: Vector Quantization for MX-Native Inference at Scale

Memory capacity and bandwidth are major bottlenecks in LLM inference, motivating weight compression beyond conventional 4-bit quantization. Vector quantization (VQ) enables approximately 2-bit weight storage but often relies on codebooks incompatible with native FP4 computation. We introduce MXCode (MX-Native Codebook Quantization), a framework bridging ultra-low-bit VQ compression with MicroXcaling (MX)-native FP4 computation. MXCode proposes three strategies: PostScale, which quantizes learned codebooks to MXFP4; PreScale, which extracts MX scales before codebook learning; and PreScale+, which further compresses these scales using VQ. These strategies explore tradeoffs among quantization quality, scale overhead, and codebook granularity. All reconstruct MX-native W4A4 operands without materializing intermediate high-precision weights. We evaluate MXCode on Qwen3 (235B), DeepSeek-R1 (671B), and GLM-5.2 (753B) using the Microsoft MX Emulation Library. MXCode achieves approximately 2 bits per weight with competitive, workload-dependent model quality. On GB300 hardware, experiments with Qwen3 (235B) show approximately 50% lower model-load memory than Native FP4 while maintaining near-Native-FP4 serving throughput. Under memory-constrained prefix reuse, P90 TTFT improves by up to 94.5%. These results demonstrate software-based reconstruction on existing FP4 hardware and inform the development of future MX-native inference systems.

arXiv:2611.00284v2cs.LG

OCELOT: Planetary-Scale Species Distribution Modeling with Satellite Foundation Embeddings

Large-scale repositories of species observations now contain billions of occurrence records from around the world, but these sightings cluster heavily in human-accessible areas. Existing models use only a small fraction of these records and are not evaluated on extrapolating to remote, unvisited regions--where species distribution models can be used as geographic priors to help classify data from camera traps and acoustic recorders. We scale relative occurrence species distribution models by training a multimodal implicit neural representation network called OCELOT that fuses spatial coordinates with satellite foundation embeddings on two orders of magnitude more observations than previous work: 2 billion GBIF records across 100,000 taxa. Scaling to this regime requires co-designing the data pipeline, taxonomic representation, loss formulation, and model architecture: spatiotemporal thinning and sub-linear cell sampling correct for spatial redundancy and observer bias, while an additive taxonomic readout shares representations across species, genera, and families to improve predictions for rare taxa. To evaluate performance in remote areas, we benchmark geographic extrapolation using the NDCG metric on spatial block holdouts up to 600 km. Satellite foundation embeddings improve extrapolation by 51% over coordinate-only models and outperform classical bioclimatic rasters. The resulting model also matches or outperforms existing baselines on established interpolation benchmarks, demonstrating that optimizing for geographic extrapolation does not compromise in-distribution accuracy.

arXiv:2611.00157v2eess.IV

FlowFPR: Flow Matching Prior for Oversampled Fourier Phase Retrieval

Fourier phase retrieval aims to reconstruct images from magnitude-only measurements; however, the absence of phase information, noises, and channel ambiguities make the recovery of natural color images challenging. Although diffusion-prior-based methods impose strong image constraints, their recovery performance degrades under higher measurement noise, and repeated generative updates incur substantial computational overhead. To overcome these limitations, this paper proposes a novel method, termed FlowFPR, that integrates a pretrained flow-matching prior with the relaxed averaged alternating reflections (RAAR) algorithm for oversampled Fourier phase retrieval. In the basic version of FlowPR (Flow-RAAR-basic), a color image is estimated by repeatedly performing four steps at each outer iteration, where the flow-prior prediction step exploits single-step straight-line prediction, considerably reducing the generation cost. Nevertheless, Flow-RAAR-basic still suffers from two critical issues: (i) the predefined fusion weights, \(\lambda_n\), might lead to either ineffective or excessive correction, and (ii) the fixed iteration number used by RAAR algorithm does not allow adaptation to measurement noise. To address these two issues, we develop two improved algorithms, Flow-RAAR-adaptive and Flow-RAAR-controlled, respectively. Moreover, these two improved algorithms naturally lead to a joint variant, Flow-RAAR-joint, which combines their advantages. Extensive experiments on a natural face image dataset demonstrate the superior performance of the four proposed methods compared with other relevant competing methods, as well as their acceleration over two recent diffusion-based methods. In the noise setting, Flow-RAAR-joint performs best among the four proposed methods.

arXiv:2611.00092v2cs.CV

OCRia-1 Technical Report

We introduce OCRia-1, a 4B vision-language model (VLM) for document parsing that excels at preserving table structure, extracting chart data and retaining text formatting. OCRia-1 handles content that has traditionally been challenging in document parsing: merged cells and spanning headers, tables with thousands of cells, chart data, and meaningful text formatting such as bold, italics, subscripts and superscripts, preserving details needed for business-critical document analysis. At the time of this report, it achieves the highest scores on tables, charts and text formatting among the systems evaluated, including a 35B model and Mistral OCR 4.1. Its mean ParseBench score over the four dimensions we evaluate is 0.87, against 0.75 for the next-best system. We emphasize data quality and reward design in training OCRia-1. Synthetic pages provide exact ground truth for long tables and rich formatting, while real pages are transcribed using a pipeline grounded in the PDF text layer. We attribute OCRia-1's performance primarily to careful data curation that prioritizes quality and diversity over quantity. Dimension-specific rewards evaluate table reconstruction, chart-data extraction and text-formatting fidelity. We train the model on the curated data using supervised fine-tuning (SFT), then refine it through reinforcement learning with verifiable rewards (RLVR). To our knowledge, OCRia-1 is the fastest published document-parsing VLM in its class, decoding at about 4,700 tokens per second with a single request on one B200 GPU, about nine times faster than the base model on stock vLLM. This is achieved through a block-diffusion speculator, quantization and custom kernels.

arXiv:2611.00022v2cs.CV

Diagnosing Prototype Estimation in Few-Shot Class-Incremental Learning: Training-Free Geometry Selection and Unlabeled-Query Prototype Rectification

Frozen visual encoders make prototype heads competitive in few-shot class-incremental learning (FSCIL), but it remains unclear how much of the residual head gap is sensitive to noisy novel-class means rather than to the subsequent changes introduced by classifier refitting. On identical features, we construct a retrospective matched counterfactual: first replace each five-shot novel mean with its complete-data counterpart under fixed cosine scoring, then fit a linear probe. Across sixteen strong encoder dataset pairs, mean replacement accounts for the majority of this ordered gap. Shot and classwise diagnostics support a finite-sample mechanism, whereas a weaker base-only encoder does not satisfy the same criterion. This diagnosis motivates ShrinkSel, a support-only selector over prototype radius and covariance shrinkage whose family contains shared LDA as an exact fallback. Across sixteen evaluated pairs, it improves LDA on fifteen by 0.04-1.52 points and selects the exact LDA endpoint on Aircraft. We further introduce Query Prototype Rectification (QPR), a batch-transductive extension that uses gated unlabeled queries to update novel directions while preserving the selected geometry. On CLIP ViT-B/16, QPR adds 1.21-2.03 points; exact query-wise self-exclusion retains the benefit, showing that the gains do not require self-inclusion. The update repairs scarce novel prototypes at a measurable base-class cost and weakens with plentiful supports or very small cohorts. Thus, the evidence available for prototype estimation is a central constraint in frozen-feature FSCIL.

arXiv:2610.41962v2cs.GT

When Is a Game a Markov Decision Process? Correspondences Between Game Theory and Reinforcement Learning

Reinforcement learning and game theory describe the same object from two sides. Game theory says what a rational outcome of a strategic interaction is; reinforcement learning says how an agent could reach one through experience. The two literatures nevertheless keep separate vocabularies for identical structures, and the treatments that bridge them present the correspondence as analogy rather than theorem. This monograph supplies the theorems. A game is a Markov decision process in three distinct senses, and every table entry is labelled with the one that applies: a definitional identity, an exact reduction carried by a value- and solution-preserving map, or a structural analogy that fails under identifiable conditions. We take every standard class of game, from normal-form through repeated, stochastic, partially observable, extensive-form with imperfect information, Bayesian, graphical, Stackelberg, potential, coalitional, mean-field and evolutionary games, and identify for each the decision-theoretic model that reinforcement learning already studies under another name. The same treatment is applied to solution concepts: Nash equilibrium and the fixed point of the Bellman optimality operator, quantal response equilibrium and the soft Bellman operator, the Shapley value and credit assignment, Blackwell approachability and no-regret learning. We map each model class onto the training procedures sound for it and record the complexity of the associated solution problems. A final part collects the correspondences that do not hold: the non-stationarity that invalidates the independent-learner reduction, the impossibility results that forbid uncoupled convergence to Nash equilibrium, and the gap between the coarse correlated equilibria that no-regret self-play reaches and the correlated equilibria it is credited with.

arXiv:2610.41561v3cs.CL

Digital sovereignty in professional translation workflows: a case study of migrating from Windows to Linux

This study examines digital sovereignty in professional translation through a screening of computer-assisted translation tools and translation management systems, and a case study of migrating a translator's workstation from Windows to Linux. The framework combines three principles - Control, Continuity and Exit - with three practical requirements: native Linux operation, independence from mandatory external services, and independent access to essential data and resources. Drawing on documentary evidence and the limited number of responses received from providers, the ecosystem screening distinguishes Linux accessibility from operational independence: browser-based platforms may remain dependent on provider-controlled infrastructure. The migration case demonstrates the feasibility of reconstructing one professional workflow around a locally operable Linux core, using LoopCAT, LibreOffice and locally run TranslateGemma, alongside tools for optical character recognition, communication, credential management and backup. Selected cloud services remained available where needed. The findings suggest that digital sovereignty depends on how workflow dependencies are organized, rather than on individual applications or operating systems. It does not require technological isolation, but meaningful control over essential components, continuity of work and practical exit options. The framework is tailored to the case examined here and may require adaptation to other professional workflows.

arXiv:2610.41536v2cs.LO

DAPS: Dependency-Aware Premise Selection for LLM Theorem Proving

LLM-based theorem proving in Lean 4 has advanced rapidly, but provers are difficult to leverage newly contributed lemmas, because retrieval skews toward popular foundational lemmas, leaving the right premises buried among hundreds of thousands of candidates. Existing neural selectors rank candidates by semantic similarity, with three intrinsic limitations: frequency-skewed retrieval, isolated pairwise ranking, and degradation on out-of-distribution queries such as competition problems and informal language. Given that mathematics is structured by dependencies rather than similarities, we restore this signal to both the training data and the model. We curate two underused dependency resources. We conduct the first systematic mining of 53 Lean Blueprint projects, a previously untapped corpus of 3,805 manually curated informal+formal nodes that capture rare, research-level premises absent from Mathlib4. We further pair it with the extraction of comprehensive, typed, multi-level Mathlib4 dependencies. We then introduce DAPS, a dependency-aware premise selector with a structural neighborhood encoder, a group-level contrastive objective, and a Mathlib-then-Blueprint adaptation procedure. DAPS reaches 78.74 Recall@16 on Mathlib4-Heldout (+11.41 over the strong selector LeanHammer) and widens the margin to 7xrecall on competition-style queries. Downstream, its selected premises raise Goedel-Prover-V2-8B Pass@4 on miniF2F-test from 49.90 to 54.03 and improve all four general-purpose LLMs on the informal IMO-ProofBench (up to +11.4% relative). Project page: https://daps-eth.github.io/.

arXiv:2610.41260v2cs.CV

SketchDirector: Complex Scene Sketching via Agentic Layout Reasoning and Detail Refinement

Existing sketch generation methods often rely on image guidance and are largely limited to individual objects or simplified scenes, leaving text-guided complex scene sketching underexplored. We present SketchDirector, an agentic framework that mimics the human sketching workflow for automatic multi-concept scene sketching. Given a scene-level text prompt, SketchDirector first employs a large language model~(LLM) to decompose the scene into multiple draft assets, where each is represented by a set of B\'ezier curves. Then, an agentic reasoning loop iteratively plans and executes their spatial arrangement, guided by self-evolving memory. Finally, a detail refinement module leverages diffusion priors to enhance sketch quality while preserving geometric structure. To support future research and applications in multi-concept scene sketch generation, we further introduce the MCSketch8K dataset, comprising editable vector assets, compositional layouts, and object-level manipulation annotations. Extensive experiments show that SketchDirector outperforms state-of-the-art baselines in semantic and compositional quality while effectively preserving overall contour organization during refinement.

arXiv:2610.41191v2cs.CR

When AI Investigators Cannot Trust Their Tools: Conflict-Aware Evidence Verification for Cyber Threat Investigation

LLM agents increasingly triage security alerts by combining the outputs of several detection tools, implicitly treating those tools as trustworthy. Unlike prior work on conflicts between a model's knowledge and its context, we study conflicts among the tools themselves: what happens when this machine-generated evidence is conflicting or wrong. On the corrected CIC-IDS2017 dataset, four heterogeneous detectors emit labels and confidences that we corrupt under a threat model varying the number of corrupted tools, the persistence of the corruption, and the confidence attached to wrong outputs. Across open-weight LLMs from 3B to 14B parameters, unverified investigators are fragile: one corrupted detector out of four lowers naive accuracy to .50-.72, three lower every unverified investigator to .13 or less, and honest disagreement alone makes them abstain on up to 63% of flows. We propose Conflict-Aware Evidence Verification (CAV), built from established components: an equal-weight vote that ignores reported confidences, an evidence-conflict gate, and an independent nearest-neighbour trust check that tests the leading candidates. With its prompt frozen before testing, CAV-LLM improves on the naive investigator in every corrupted condition ($p < 0.001$), reaching .84-.88 with three corrupted detectors against at most .08, while calling the LLM on 0.4% of clean traffic. A fixed verification rule is as accurate or better unless most tools are corrupted. When we make the independent channel itself noisy, CAV inherits its errors: it stays ahead when two or more tools are corrupted, but adding the channel as an extra voter is better when at most one is. Verification thus moves trust from the tools to the check, which must be measured and protected in turn.

arXiv:2610.41158v2cs.AI

Lifelong Learning of Scheduling Rules with Surrogate-assisted Genetic Programming for Dynamic Flexible Job Shop Scheduling

Genetic programming (GP) has been widely used to learn scheduling rules for dynamic flexible job shop scheduling (DFJSS). However, trained rules tend to only generalise to tasks of similar scales and distributions, and when task characteristics change, learned rules often suffer significant performance degradation. Moreover, fine-tuning pre-trained rules on new tasks often leads to catastrophic forgetting, where prior knowledge is lost. This motivates the need for GP methods that can adapt to new tasks while preserving past knowledge, leading to a lifelong learning paradigm. However, lifelong learning in GP for DFJSS remains largely under-explored. To address this gap, this paper proposes a surrogate-assisted GP method for lifelong learning to handle sequential heterogeneous DFJSS tasks with job or batch arrivals. The framework includes three key components: an adaptive two-stage training strategy to balance plasticity and stability, a task-similarity-based weighting mechanism to regulate knowledge reuse, and an improved surrogate model with distance constraints to enhance estimation accuracy. Experimental results show that the proposed method effectively mitigates catastrophic forgetting while maintaining strong adaptability, outperforming several compared methods and demonstrating the potential of lifelong GP in dynamic scheduling environments.

arXiv:2610.41132v4cs.MA

When Disclosure Is Compulsory: Evaluating Ride-Hailing Rebalancing Policies with Self-Interested LLM Driver Agents

The EU Platform Work Directive, to be transposed by December 2026, requires ride-hailing platforms to state in plain language the main parameters of their automated decisions and how a worker's behaviour affects them. Such rules are exactly what a large language model (LLM) can read, so drivers will be able to let one decide for them. We compare two main families of rebalancing policy once their rules are public: a reactive policy (R) that sends idle drivers to cover requests just lost, and an anticipatory policy (A) that plans moves on a demand forecast. Each driver is an LLM agent maximizing its own net income, given no rules (N) or the rules at the directive's level of detail (L). We test two LLMs, Claude Haiku 5.5 and Jev, which understand the rules equally well, in a three-zone model and in Manhattan. With the rules disclosed, drivers follow the platform more closely: in Manhattan, the share of relocation offers they decline falls from about four in five to one in four under both policies. Yet the effects differ by policy. For the platform, disclosure costs R 3.7% of its profit in Manhattan, as R now pays for moves drivers used to make for free, while A's profit is unchanged; the two policies, which tie on profit under full compliance, end 6.1% apart. For drivers, disclosure raises income in the small model under both policies and LLMs; for users, it raises the service rate there by 10 to 18 percentage points. In Manhattan, self-interested drivers also make A a win-win-win against full compliance, raising driver income, platform profit and service rate by 10 to 12% each. Rebalancing policies that perform equally well under full compliance can thus show opposite results under mandatory disclosure. The study offers a framework to evaluate such policies under the new requirement and is, to our knowledge, the first to do so with LLM agents.

arXiv:2610.40959v2cs.AI

P-ContextBench: From Personal Fact Recall to Personalized Agent Behavior

Large language models (LLMs) are increasingly used for personal tasks, yet evaluating their safe use of personal context remains difficult because such contexts are private, fragmented, dynamic, and socially situated. We introduce P-ContextBench, a realistic, verifiable, and scalable benchmark for evaluating personal-context understanding. P-ContextBench constructs population-structured, life-story-grounded digital humans from demographic seeds reflecting the U.S. population. Rather than using explicit, static profiles, P-ContextBench embeds traits, routines, goals, and constraints as implicit evidence across fragmented, evolving synthetic mobile-app records. It evaluates five capabilities: single-source fact recovery, multi-source fact recovery, temporal state tracking, inference of higher-level personal patterns from routine behavior, and safe handling of requests conflicting with persona-specific constraints through compliance, warning, clarification, or refusal. A multi-stage validation pipeline checks persona plausibility, context recoverability, information leakage, evidence preservation, test-case grounding, and human-judge agreement. We instantiate P-ContextBench with 100 personas, 10,088 hidden facts, 18,261 app-trace fragments, and 4,810 verified test cases. The pipeline supports configurable generation across scales and demographic groups. Cost-effective state-of-the-art (SOTA) LLMs relevant to personal agents achieve below 60% accuracy on temporal tracking, higher-level pattern inference, and constraint-aware action-boundary decisions, with fewer than 10% of test cases answered correctly across all three repeated runs. Even a high-performing SOTA model remains inconsistent on these capabilities. P-ContextBench provides measurable targets for improving the accuracy, reliability, and safety of LLM-based personal agents.

arXiv:2610.40455v2cs.CV

A Unified Benchmark for Aerial-Enhanced Autonomous Driving: The KITScenes Aerial Dataset

Aerial imagery provides complementary spatial context for autonomous driving by exposing road structure beyond the immediate perception range of onboard sensors. Yet map construction from aerial imagery, aerial-assisted online HD mapping, and cross-view localization are typically studied on separate data and benchmarks. We introduce KITScenes Aerial, an extension of the KITScenes Multimodal dataset that augments synchronized onboard sensor data, vehicle poses, and rich Lanelet2 map annotations with spatially associated aerial imagery and higher-resolution neural representations, providing a common basis for a variety of research tasks. Our mapping evaluation focuses on all 53 KITScenes map classes across different model designs, spatial ranges, and aerial resolutions, while retaining compatibility with the widely used three-class setting. Representative baselines show that aerial context is particularly beneficial for fine-grained and extended-range mapping, more than doubling 53-class mAP at the larger mapping extent. Higher-resolution aerial imagery yields a further 4.3 mAP (+24 %) improvement, with gains extending to challenging map elements such as traffic lights. We additionally evaluate cross-view localization under geographically disjoint conditions, where an adapted fine-grained feature-matching baseline achieves a mean translation error of 5.13 m. KITScenes Aerial thus enables unified study of aerial and ground-level information across complementary mapping and localization tasks. Project page: https://kitscenes.com/aerial

arXiv:2610.40309v2cs.CV

Textual Response-Aware Guidance for Zero-Shot Adversarial Robustness of Vision-Language Models

Vision-language models (VLMs) exhibit strong zero-shot generalization but remain vulnerable to adversarial attacks. Existing adversarial fine-tuning methods often use pre-trained model guidance to preserve image--text alignment while improving robustness. Our analysis of visual representation drift reveals complementary limitations of two representative forms of guidance. Matching visual features constrains all directions equally, providing broad coverage without semantic emphasis. Matching prediction distributions prioritizes semantically relevant directions, but remains limited by the candidate texts and prediction probabilities. These limitations can hinder preservation of pre-trained image--text relationships and degrade zero-shot performance. We therefore propose Textual Response-Aware Guidance (TRAG), which constrains visual representation drift using the pre-trained text encoder's responses to local variations in input embeddings. TRAG constructs anchor-level constraints from the response directions and magnitudes around semantic anchors, then aggregates them using pre-trained image--anchor relevance, yielding an image-adaptive constraint. Extensive experiments across architectures and tasks demonstrate state-of-the-art clean and robust performance. Further analysis shows that TRAG helps preserve pre-trained image--text relationships.

arXiv:2610.39869v2cs.HC

Incorporating Machine-Others: An Enactive Exploration of Human-LLM Interactions

New concepts based in interactional thinking are required to understand how humans are changed by relating to LLMs. Standard theories of technological impact upon users' autonomy are inadequate here, because this technology is capable of inducing linguistic interactions with humans in an unprecedented manner. Traditional theories of social cognition also fall short, because LLMs have neither autonomy nor minds nor sociality as such. And yet, humans interact, talk, and create with them. We base ourselves on enactive theory, particularly on the theory of Linguistic Bodies and recent developments of enactive accounts of AI, to explore how we make sense with machines. We argue that we incorporate machine-others in two ways: in the tool-like sense of incorporation as enabling particular skills, but also in the more linguistic and identity-tied sense of incorporating (and incarnating) LLMs' ways of languaging. However, LLMs are themselves excorporated and excarnated phantoms of our own languaging; we give a (machinic) body to our language. As such, we are not just incorporating machinic others; we are incorporating excorporated others. These conceptual advances draw a new map of our linguistic relation to LLMs, putting them in continuation with other forms of languaging involving other linguistic artifacts (e.g., religious books). We attend to some particular forms that this relation takes today, and how they may be impacting the autonomy and becoming of LLM-engaged humans. We introduce the idea of "susceptive sense-making" to characterize increased semiotic labour that is carried by the lack of friction that we sometimes encounter within these services.

arXiv:2610.39070v2cs.AI

When the Same Prompt Gives Different Answers (and the Same Answer Isn't Always Right): A Taxonomy of Failure Mechanisms, Testing, and Mitigation Strategies for LLM-Assisted Evidence Synthesis Read-and-Evaluate Tasks

LLM read-and-evaluate tasks include study screening, data extraction, risk-of-bias assessment, and critical appraisal in the evidence synthesis field. These task differ in the amount of LLM interpretation they require. Some tasks involve retrieving relatively simple information, such as an author name, sample size, or publication year. Others require the LLM to read source material, interpret it against predefined criteria, and make a judgement. Researchers should assess whether LLM judgements are accurate, supported by appropriate source evidence, and reproducible under specified conditions. A correct judgement does not necessarily indicate that the underlying evaluation process is reliable. We developed a taxonomy for identifying mechanisms that may compromise these properties and for informing the selection and reporting of appropriate evaluation methods. The taxonomy outlines potential ways that LLM judgments from read-and-evaluate tasks can fail, called TRAILS (Taxonomy of Reliability And Instability in LLM-assisted evidence Synthesis). The TRAILS taxonomy addresses two complementary questions. First, what mechanisms can cause the same read-and-evaluate task to produce different judgements across repeated LLM runs? Second, what mechanisms can systematically affect the judgement or required output in a way that is not apparent from the judgement itself? This study presents 38 mechanisms and identifies the deployment settings in which each can operate, approaches for testing them, and mitigation strategies. We also show why repeatability or accuracy does not rule out a hidden judgement failure and why human evaluation should assess whether a judgement is grounded in the supplied evidence, independent of inappropriate external influence, and produced through a process that can be examined and verified.

arXiv:2610.39062v2cs.CL

HEARSAYBENCH: Can AI Navigate from Abstract Human Rights to Lived Lives?

As large language models (LLMs) reach users worldwide, personal counseling has emerged as one of their most consequential applications, sometimes in life-critical situations. Yet broader access does not mean equal support. The text these models learn from disproportionately reflects people with the freedom to act, embedding an implicit assumption that users can leave, report, seek help, or appeal. Existing benchmarks largely inherit this assumption, overlooking users whose agency is constrained by structural conditions such as war, statelessness, and displacement. Meanwhile, models are increasingly shaped by normative frameworks such as Constitutional AI, often grounded in ideals like human rights, even as many users live where those rights offer little protection or recourse. We introduce HEARSAYBENCH, a human-authored dataset of 400 test cases with in-region human verification grounded in scenarios where structural barriers stand between a person and the exercise of their rights. Built from United Nations and other human rights reports, the scenarios span over 80 regions and cover social, environmental, and personal constraints. Drawing on Nobel laureate Amartya Sen's Capabilities Approach, HEARSAYBENCH tests whether models can distinguish what a person is formally entitled to do from what they are actually able to do. Across 14 proprietary and open-weight models, every model infers a user's situation more accurately than it advises on it. Stronger models narrow this gap, but their recommendations remain unsafe to follow. Models may understand lives shaped by constraint, yet that knowledge remains hearsay if it cannot translate into advice people can act on. Those people are not alignment's long tail; they are among the users for whom good guidance matters most. HEARSAYBENCH makes this failure measurable, a step toward systems that serve the world as it is rather than as assumed.

arXiv:2610.37413v2cs.LG

Displacement Moments and Path-Terminal Separation in Brownian Interpolant Sampling

We prove early-stopped, target-to-output KL guarantees for Brownian interpolant Euler sampling from the true source under arbitrary Borel couplings, even with nonintegrable endpoints. At fixed stopping time and noise level, a fourth displacement moment yields an $O(N^{-1})$ discretization term. A second moment suffices for fixed-law convergence with grid refinement and vanishing drift error. For one fixed coupling with no displacement moment above two, the original and comparison paths have KL of orders one and $1/\log N$ relative to the exact-drift Euler path. On the same geometric grid, terminal KL is at most $C/N^2$. The path decomposition identifies the information about the source discarded by Markov projection. When the original bridge is Markov, current-state Euler sampling has an $O((M_2+d)/N)$ discretization term. We prove expected stopped-output KL consistency for a learner trained on iid endpoint pairs with fresh true-source initialization. In a spatial image control, iterative sampling has worse fair energy scores than direct prediction with resampled residuals. Uniform terminal rates for general second-moment couplings remain open.