Artificial Intelligence · Cross-listed from cs.CR
MutMem: Cryptographically Authorized Mutation in Persistent Agent Memory
Walid Saidi
Persistent
agent memory must adapt as later outcomes change earlier evidence, yet mutable
retrieval weights create an attribution problem: reviewers must distinguish authorized adaptation from database tampering. We present MutMem, an authorized-mutation protocol in HOM-AIMOS, a persistent
agent-
memory engine. MutMem retains
memory content, records signed positive and negative outcome evidence without age-based expiry, and commits each nontrivial weight change as a housekeeper-authorized transition. Each transition binds a terminal provenance node, signer epoch, quantized old and new weights, a no-fork predecessor, and two domain-separated SHA-256 commitments. Ed25519 verification runs in both the database writer and a portable verifier. Content classified as poison-likely is retained with signed, revisable labels used by recall as trust evidence. We evaluate utility, mutation integrity, and poisoning adaptation. HOM-AIMOS answers 459/500 LongMemEval questions correctly under LLM judgment (91.8%). On LoCoMo, it obtains 74.12% judged accuracy and, under a separate upstream-compatible protocol, 58.20
token F1. A native suite passes all declared authorization,
topology, tamper, signer-epoch, and post-mutation-recall cases; median signed-transition latency is 4.865 ms. In a declared N=100 PoisonedRAG adaptation, no injected poison appears in attacked top-5 disclosures (0/100; 95% Wilson upper bound 3.70%), while induced target-answer attack success among 98 clean-negative targets is 1/98 (1.02%). A preregistered four-arm ablation attributes the
retrieval reduction to signed stored labels: the retriever selects poison for 94/100 targets when epistemic policy is bypassed and 0/100 when labels are restored. MutMem provides evidence of integrity, authorization, traceability, and historical continuity; it does not establish content truth.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
CURV: Enhancing Chart Understanding Through Curriculum Visual Grounded Reasoning
Xuehang Guo, Pingyue Zhang, Ruiyi Zhang et al. (+6 more)
Chart question answering (CQA) requires
multimodal large
language models (MLLMs) to integrate visual comprehension with logical
reasoning, yet current models struggle with accurate visual grounding and coherent
reasoning chains. While extrinsic chain-of-thought prompting and visual cues significantly improve performance, current MLLMs lack intrinsic visual grounded
reasoning capabilities, leading to inaccurate perception and
reasoning disconnected from visual evidence. To address these limitations, we propose CURV, a curriculum learning framework that develops intrinsic visual
reasoning capabilities by reformulating CQA as multi-step visual grounded
reasoning, where each step coordinates logical
reasoning with dynamic visual grounding through spatial
attention concentration. To assist model learning, we further introduce CCQA, a three-level curriculum
dataset with scalable synthetic generation across diverse chart types and
reasoning patterns. Our curriculum systematically progresses from basic single-operation
reasoning to complex multi-chart compositional tasks. Experiments demonstrate that CURV achieves up to $\uparrow20.50\%$ improvements over baselines and is generalizable to real-world
benchmarks (up to $\uparrow12.30\%$) and out-of-domain
multimodal reasoning tasks (up to $\uparrow10.20\%$), validating the effectiveness of internalizing visual
reasoning with dynamic grounding for enhanced chart understanding capabilities. Code is available at: https://xhguo7.github.io/CURV/.
Read the full abstract ▾
All authors Xuehang Guo, Pingyue Zhang, Ruiyi Zhang, Zhenhailong Wang, Hanrui Lyu, Heng Ji, Tong Sun, Qingyun Wang, Manling Li
Artificial Intelligence · Cross-listed from cs.CV
In-Context Collapse in Vision-Language Models and How to Mitigate it?
Mohammad Rostami
Many-shot in-context learning (ICL) lets
vision-
language models (VLMs) adapt from image–label demonstrations without weight updates, and is widely assumed to improve as more demonstrations are supplied. We show the opposite: as demonstrations accumulate, a subset of VLMs undergo an in-context collapse, a sharp, sometimes catastrophic accuracy drop spanning synthetic classification, natural-image classification, and VQA
benchmarks, in some models falling below chance while outputs remain well-formed. Across an open VLM panel ($0.5$B–$11$B) and a frontier model (Claude Sonnet 4.5), the collapse is graded. Two capabilities turn out to be dissociable: robustness to accumulating demonstrations and the ability to learn a novel rule in context, their combinations yield three reproducible regimes. A parameter-matched lesion-and-rescue causally localizes the collapse to the
vision-
language integration pathway: an adapter on the connector and early/mid layers restores genuine learning (remap accuracy $0.39!\rightarrow!0.91$ at 16 shots), while an equal-capacity adapter on the late readout does not. We propose CircA, whose core is a one-time integration vaccine: trained once on one synthetic task, it transfers collapse-resistance to unseen task families (chance$\rightarrow$$0.71$/$0.60$ on CIFAR/Fashion). The layers best for in-context integration are not the layers best for weight-based consolidation, the late readout achieves higher accuracy and less forgetting at fewer parameters. The collapse is an integration failure at the
vision–
language interface, correctable by a lightweight, transferable intervention.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Wiring Beats Blending: Structure-Aware Compensation for Transformer Downscaling
Ravi Satya Durga Prasad Yenugula
Model families are trained size by size. Can a pretrained large model instead be converted into a smaller sibling? We characterize the 1.4B->410M conversion in Pythia end to end. Representations align strongly across sizes (ridge R^2=0.84); parameters align weakly. Dense weight projection is destructive; a bit-exact control places the fault in basis mixing, which breaks rotary, per-head, GELU, and LayerNorm structure. Residuals after the best-fit linear operator carry no learnable or transferable signal under shuffle controls, so conversion value lives in initialization. Matched-budget continued pre-training separates two independent levers, least-squares compensation (function lever, best zero-shot) and variance-preserving rescale (dynamics lever, best endpoints), with a placement rule read off the architecture: compensation is well-posed exactly where no normalization sits between cut and read; norm-fronted paths take rescale. At 30M
tokens compensation beats the best subcloning variant on a width-reduced pair (84.6+-2.1 vs. 90.7+-3.8, 6/6 seeds) and a held-out depth-reduced pair (109.3 vs. 117.9, 3/3 seeds); the two reach parity at 33x the budget, and at ~5x the donor scale the levers anti-synergize. A second family, Qwen3-1.7B->0.6B (RMSNorm, SwiGLU, grouped-query
attention), takes the placement rule unchanged but reverses a 2x2 crossing the re-fit with the activation statistic it reads: on Pythia the re-fit carries the gain and the statistic adds 2.3 perplexity; on Qwen3, with far more heterogeneous MLP unit energies, the statistic carries 66 points and the re-fit is a null on top. Informed selection + compensation + rescale is the best arm in both families, 19x and 23x lower perplexity than from scratch at matched budget (2.94 and 3.13 nats), and beats structured
pruning plus
distillation. Code, checkpoints, and the frozen eval corpus are released.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from quant-ph
Improved Quantum Algorithms for Reinforcement Learning Under a Generative Model
Joao F. Doriguello
Reinforcement learning is a subfield of machine learning that studies how an
agent interacts with an environment in order to extract as large a reward as possible. A standard approach to study such interaction is through Markov Decision Processes (MDPs) and the task of choosing an optimal policy — a function that tells the
agent which action to take. In this work, we study two types of MDPs — finite-horizon and infinite-horizon discounted — and propose new
quantum algorithms for computing approximate optimal policies. Our
quantum algorithms are based on a new combination of standard value iteration and
quantum subroutines like
quantum mean estimation and
quantum maximum finding, overall enhanced with techniques from sample-optimal classical
algorithms. Our resulting query complexities improve upon previous works, thus approaching already established
quantum lower bounds.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CR
Evading Chain-of-Thought Monitoring Through Model Poisoning
Giorgio Severi, Shujaat Mirza, Blake Bullwinkel et al. (+1 more)
Chain-of-thought (CoT) monitoring is an increasingly important component of AI safety stacks but relies on the assumption that a model's
reasoning trace is informative about its actions. This work studies the limits of CoT monitoring through the lens of model poisoning. We demonstrate that
backdoors can be implanted into
reasoning models to elicit an attacker-chosen behavior while their CoT traces appear entirely benign. We find that these CoT-Hidden
backdoors can be induced through simple fine-tuning recipes across
reasoning-model architectures and sizes. When direct poisoning is ineffective, we introduce a curriculum training approach that progressively teaches the model to produce an attacker-chosen output while concealing the behavior from its
reasoning traces. These findings suggest that CoT monitoring may be better framed as a question about the consistency between a model's
reasoning trace and its final response than as anomaly detection within a trace. We further examine the mechanisms that allow models to suppress evidence of the target behavior from their
reasoning traces. Causal interventions locate a trigger-conditioned activation pathway that does not depend on the visible
reasoning, and residual stream verbalizations provide an anomaly warning near answer generation, but do not identify the trigger, target, or
backdoor mechanism.
Read the full abstract ▾
All authors Giorgio Severi, Shujaat Mirza, Blake Bullwinkel, Amanda Minnich
Artificial Intelligence · Cross-listed from cs.CL
Learning a Vector-Symbolic Model for Socio-Cultural Tasks
Meera Ray, Swapnika Dulam and Christopher L. Dancy
How can we better represent the impact of sociocultural structures on decision making in computational cognitive models? Modeling this impact requires traversing multiple levels of semantic representation, however it is not immediately clear to a modeler which levels of representation are most salient to a given situation. Though large
language models and cognitively grounded corpus models can represent broad semantic associations through co-occurences, the role of self representations in
memory should be accounted for to determine how cultural associations shape decision making. We propose a declarative
memory system to be used in the ACT-R cognitive architecture that represents semantic associations at multiple levels via a vector-symbolic autoencoder. We use a simple HRR operation to encode episodic
memories differently from semantic
memory vectors extracted from text to produce a final chunk activation for a
memory request. We use ACT-R cognitive models of a racially contextualized implicit association test (IAT) to test this new declarative
memory system.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
A Unified 2D Framework for DeepLesion Detection, Segmentation and Short Report Generation
Ruida Cheng, Tejas S. Mathai, Benjamin Hou et al. (+4 more)
In previous work, we integrated large
language models (LLMs) into the lesion
segmentation model based on the ULS23 DeepLesion
dataset, using short-form findings from the reports. In this study, we developed a unified 2D lesion analysis framework that integrates LLM-based
reasoning, lesion bounding box detection,
segmentation, and radiology report generation from the original DeepLesion
dataset. In the testing phase, we achieved relatively high lesion bounding box detection accuracy with mAP50 of 70.1%, mAP50-95 of 46.4%; Lesion
segmentation performance with a Dice score of 62.6%; short report generation accuracy with BLEU_1 score of 64.3%, BLEU_4 score of 49.6%, METEOR of 34.7%, and ROUGE_L of 60.1%. In this work, we address the challenging issue of
segmentation in the original DeepLesion
dataset and achieve a 28.5% Dice score improvement over the nnUNet lesion
segmentation model. We also integrated spatial and anatomical context into the DeepLesion short report generation. We released the implementation,
dataset, and models on Github. https://github.com/ruida/2D_DeepLesion_Foundation
Read the full abstract ▾
All authors Ruida Cheng, Tejas S. Mathai, Benjamin Hou, Qingqing Zhu, Zhiyong Lu, Matthew McAuliffe, Ronald M. Summers
Artificial Intelligence · Cross-listed from cs.CV
SAGE: Semantic Explainability of Attention-Based Survival Models in Computational Pathology
Abdallah Lamane, Abdul Rahman Diab, Ren-Chin Wu et al. (+1 more)
Attention-based multiple instance learning (ABMIL) is the predominant approach for slide-level prediction in computational pathology, yet its
attention maps provide only local explanations: they indicate where a model focuses but not which histological features drive its predictions or how the model behaves across a patient cohort. We present Semantic
Attention Global Explanations (SAGE), a post-hoc framework that extracts global,
language-grounded explanations from a frozen ABMIL model. Using a pathology
vision-
language model, SAGE scores image patches against a dictionary of 25 histological concepts, aggregates these scores according to the model's learned
attention, and quantifies how each concept relates to prediction risk across a cohort. Applied to survival prediction using seven TCGA cancer cohorts and three foundation models, SAGE recovered established prognostic features, such as the adverse association of necrosis, while revealing cancer-specific biology, including a favorable angiogenic signature in renal
cell carcinoma consistent with known molecular subtypes. Ablation studies demonstrated that these associations depend on the model's learned
attention rather than concept prevalence alone, and that the concept dictionary captures much of the prognostic information encoded by the foundation model features. Through semantically-grounded explanations, SAGE provides a scalable, model-agnostic framework for understanding what ABMIL survival models learn, enabling pathologists to interpret model behavior at the cohort level and offering the potential for biomarker identification.
Read the full abstract ▾
All authors Abdallah Lamane, Abdul Rahman Diab, Ren-Chin Wu, William Lotter
Artificial Intelligence · Cross-listed from stat.ML
A Hyperfinite Framework for Score-Based Generative Modeling
Sunder Ram Krishnan
Score-based
diffusion models are typically formulated using continuous-time
stochastic differential equations and measure-theoretic
stochastic calculus. In this paper, we develop a hyperfinite formulation of score-based generative modeling within the framework of Nonstandard Analysis. Starting from an internal
diffusion process on a hyperfinite grid, we derive the associated infinitesimal generator and establish its correspondence with the classical Fokker–Planck equation. We then obtain a hyperfinite backward-mean identity that yields the reverse-time drift and provides a constructive derivation of the reverse-time SDE. Building on these results, we show that minimization of an internal score-matching objective recovers the score function required by the reverse-time dynamics, thereby connecting score estimation with generative sampling directly at the hyperfinite level. Under suitable assumptions, we further derive a hyperfinite Girsanov formula and establish a relationship between likelihood optimization and Fisher-divergence objectives. Finally, we analyze the second-order consistency of the hyperfinite dynamics and show that the leading correction term depends explicitly on the fourth moment of the increment distribution, with the Gaussian value $\kappa=3$ eliminating the leading dispersion contribution. Taken together, these results provide a unified hyperfinite framework for
diffusion-based generative modeling–while laying foundations for further extensions–that links discrete grid dynamics, reverse-time
diffusion, score matching, and likelihood-based formulations within a common nonstandard setting.
Read the full abstract ▾
Artificial Intelligence
Towards a new paradigm of scientific discovery with socialized artificial intelligence
Xinjie Yao, Xingxin Xu, Xiyuan Gao et al. (+21 more)
Scientific discovery has advanced through successive transformations in the organization of knowledge. Observation and experimentation established the empirical foundations of science. Theory made it possible to derive general principles from particular phenomena. Computation extended inquiry into systems beyond direct observation, while data-intensive methods opened new spaces of pattern and prediction. Science now confronts a different frontier. The central challenge is no longer simply to produce more information, but to organize expanding knowledge,
reasoning, and evidence into a coherent process of discovery. Here, we introduce Bridging Literature,
Agents, and Zero-gap Experimentation (BLAZE), a paradigm of socialized scientific intelligence. BLAZE conceives AI not as an assistant for isolated research tasks, but as an organizational infrastructure for scientific discovery. It connects persistent knowledge, collective
reasoning, empirical validation, and human judgment within a continuous research lifecycle, transforming fragmented activities into a cumulative process of inquiry, criticism, and revision. The central premise of BLAZE is that scientific intelligence does not arise from computation alone. It emerges from the sustained interaction among knowledge, hypotheses, experiments, and collective verification. By organizing humans and machines within a shared scientific process, BLAZE makes discovery more traceable, reproducible, and cumulative while preserving human creativity, judgment, and responsibility. Socialized scientific intelligence may provide a foundation for the next era of science. Its purpose is not to replace human discovery, but to extend the scale, depth, and continuity of collective scientific inquiry.
Read the full abstract ▾
All authors Xinjie Yao, Xingxin Xu, Xiyuan Gao, Zhoupeng Guo, Kunlong Yang, Dengyu Zhao, Siqi Zhao, Zhihe Fan, Yichen Dong, Xin Li, Jiekang Feng, Jiahe Wu, Sen Wang, Beiming Yu, Kejia Zhao, Ruipu Zhao, Jiaqi Zhou, Heyang Li, Jianjun Chen, Anbo Dai, Xin Liu, Zhengtao Yu, Qinghua Hu, Pengfei Zhu
Artificial Intelligence · Cross-listed from cs.CR
Privacy-Preserving AI Verification via Minimal Information Disclosure
Sleem Abdelghafar and Gabriel Kulp
AI verification crosses a trust boundary: a verifier must learn enough to establish an authorized claim, yet the same evidence can reveal sensitive details about the model, workload, or hardware. We introduce minimal information disclosure (MID), which designs and quantifies the information content of verifier-facing evidence itself. MID measures collateral leakage with conditional mutual information: what the release reveals about the protected property after the authorized result is known. MID is general by design: it can accommodate different verification goals, protected properties, evidence sources, and deployment constraints. To demonstrate MID's practicality, we evaluate it on four physical measurements and six verification tasks spanning execution type, hardware identity, compute scale, and model identity. These experiments use three mechanism-design variables–the evidence channel, collection policy, and release transformation–but MID is not limited to these choices and can accommodate other deployable mechanisms. Across these tasks, MID produces three releases with perfect held-out verification and zero measured collateral leakage, while the remaining tasks yield explicit
privacy–utility frontiers. MID also supports ZKP-certified releases: we demonstrate our proposed linear-projection mechanism using a Groth16 zk-SNARK.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.IR
Search, Inspect, Fetch: Exploiting Structure-Aware Boolean Retrieval for Deep-Search Agents
Shuai Wang, Haodong Chen, Yu Yin et al. (+3 more)
Existing deep-search
agents use a Search-Visit workflow that retrieves whole webpages without considering the structure they expose through titles, headings, sections, and metadata. This prevents
agents from directly constraining
retrieval to parts of a webpage and often carries irrelevant content into their context. We introduce Sieve, a search-inspect-fetch strategy driven by a Boolean Query
Language (BQL): it searches webpage fields to filter candidates, uses an interchangeable ranker to order them, presents structure-rich result cards for inspection, and fetches only selected sections. Across three QA collections, Sieve is more accurate than the strongest conventional Search-Visit configuration on each collection while using 20.7-50.6% fewer
tokens. Boolean filtering improves every tested ranker, and the accuracy-context advantage persists across retriever choices and
agent backbones. Our implementation is included in the SkimSearchAgent library https://github.com/ielab/skim-search-
agent.
Read the full abstract ▾
All authors Shuai Wang, Haodong Chen, Yu Yin, Shengyao Zhuang, Bevan Koopman, Guido Zuccon
Artificial Intelligence · Cross-listed from cs.CV
Quo Vadis, World Modeling?
Yu Yang, Xuemeng Yang, Licheng Wen et al. (+17 more)
Continually improving
agents require dynamic interaction feedback beyond static supervision, yet direct real-environment interaction is costly, slow, unsafe, and hard to parallelize. World modeling offers a natural intermediate proxy that allows
agents to query lower-cost, more controllable feedback before committing to real actions. Classical world models instantiate this proxy primarily through future physical-state prediction, a formulation useful yet narrow for
agents that require actionable feedback beyond raw state transitions. In this work, we conceptualize
Agent-Centric Interactive World Proxies, shifting the fundamental paradigm from physical state transitions to
agent-usable information transitions, such as execution outcomes, retrieved experiences or skills, and verification signals, broadening the scope of world modeling to provide versatile feedback for continually improving
agents. To systematically map this design space, we organize world proxies into six functional forms based on their feedback modalities: dynamics, spatial, execution,
memory/experience, skill, and reward/verification proxies, which together characterize the primary ways world modeling serves
agent improvement. We further analyze how these proxies empower
agents across three progressive levels: L.1 Inference-Time Guidance, where proxy outputs enrich in-context information for superior decisions; L.2 Training-Time Optimization, where proxy outputs yield rewards, critiques, or synthetic rollouts for policy learning; and L.3
Agent-Proxy Co-
Evolution, where real-environment evidence continuously updates both the proxy and the
agent for co-
evolution. Ultimately, this work recasts world modeling into an
agent-centric paradigm, establishing a roadmap for building world proxies that empower
agents to plan better, learn faster, and evolve continually.
Read the full abstract ▾
All authors Yu Yang, Xuemeng Yang, Licheng Wen, Lingdong Kong, Xiaobin Hu, Dongyue Lu, Wei Chow, Xiyan Huang, Yuxiang Feng, Yue Liao, Jianbiao Mei, Daocheng Fu, Rong Wu, Pinlong Cai, Ran Yi, Ying Tai, Jiangning Zhang, Botian Shi, Yong Liu, Shuicheng Yan
Artificial Intelligence · Cross-listed from cs.SE
Don't Regenerate, Debug: A Domain-Specific Agent for Repairing Near-Miss Accelerator Kernels
Yansong Sun, Shenxiu Wu, Siyuan Chen et al. (+6 more)
Kernel generation for hardware accelerators such as GPUs and NPUs has become a proving ground for large
language models (LLMs), and state-of-the-art systems raise correctness through pipelines that couple LLMs with agentic reinforcement learning and evolutionary search. Such pipelines generate, compile, and execute large numbers of candidate kernels, discarding most of them and forgoing the opportunity to distill failures into reusable knowledge. Many discarded candidates are near-miss operators that compile and run but fail numerical validation; each embodies genuine domain knowledge and a nontrivial investment in LLM inference, cross-compilation, and hardware execution. We argue for a paradigm shift: rather than regenerate, debug. Debugging is far more constrained than generating from scratch: the search space is small and feedback is dense. We present a domain-specific debug
agent that addresses three core challenges in autonomous repair: mitigating knowledge scarcity through retrieved patterns and diagnostic instrumentation, ensuring integrity through anti-cheat detection and full-coverage evaluation, and controlling cost via convergence guards and bounded iteration. Debugging serves two complementary roles: it extends the capability frontier by recovering operators that repeated regeneration fails to produce, and it lowers cost per deliverable operator. Debug Pass@1 achieves 66.7% versus Regenerate Avg Pass@1's 25.9% and Regenerate Pass@3's 40.7%, while consuming 92.8% fewer
tokens per success than three-trial regeneration. Component ablations show that the knowledge base drives recovery, while integrity gates reject 12.5-33.3% of the successes the workflow itself accepted.
Read the full abstract ▾
All authors Yansong Sun, Shenxiu Wu, Siyuan Chen, Runlin Hou, Junhao Qiu, Junming Cao, Shudi Shao, Zhichao Lu, Qingfu Zhang
Artificial Intelligence · Cross-listed from cs.LG
Escaping Oversquashing: Addressable and Support-Aware Global Memory for Message Passing Networks
Félix Marcoccia
Virtual nodes are a natural tool against oversquashing: they replace long message-passing paths by a two-hop global route. But when many nodes share one global state, that shortcut can become a bottleneck itself. We study two properties of this global
memory. First, addressability: under constant-margin address codes and a nonlinearity that amplifies this margin, multiplicative write/read maps provide $M$ selectable
memory rows with only $O(\log M)$ address-code dimensions. Cross-
attention slots and a constrained $ELU+1$ bilinear
memory both satisfy these conditions. Second, support awareness: normalized cross-
attention has no self-key for a latent query to use as a reference. A learned private anchor supplies this reference, keeps the read bounded, and exposes the strength of the matching source mass. We demonstrate the merits of such properties on several instances of Two-Radius and Tree-NeighborsMatch: both addressable realizations solve the controlled tasks through depth $5$, where pooled VNs of comparable or larger size reach about $10.6\%$.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Can Training Logs Make Model Comparisons More Precise?
Wei-Jung Huang
Comparing stochastically trained models requires estimating both a performance difference and its uncertainty from repeated runs. We study whether training logs from those same runs can make such comparisons more precise. Because training-log
covariates are produced during training rather than measured before it, we use arm-specific
covariate adjustment: each model is adjusted only with statistics from its own runs, and the raw mean difference remains the reported effect. In a
vision study spanning three architectures and three
datasets, simple adjustments based on early training logs often reduce uncertainty in model comparisons. The main limitation is
covariate selection. Broadly searching the log pool for the most correlated statistic often adds more noise than it removes, even when useful statistics exist in hindsight. Training logs therefore appear useful for more precise model comparisons, but only when the adjustment avoids large selection noise.
Read the full abstract ▾
Artificial Intelligence
Predictive Set Theory: A Generative Framework for Cognitive Architecture with Operationalized Core Mechanisms
Yiyang Yu
Predictive processing theories portray the
brain as a hierarchical prediction engine that minimizes prediction error, yet they lack operational definitions for the structure of a "prediction," the standardized response to a prediction error, and the mechanism that maintains consistency across successive updates.
Bayesian cognitive science attempts to subsume all uncertainty under probabilistic belief updating, but it presupposes a closed hypothesis space and provides no generative account of how the objects over which probabilities are distributed become discrete, identifiable referents in the first place. This paper introduces Predictive Set Theory (PST), a formal generative framework that reconstructs cognitive architecture from first principles. PST anchors cognition in a minimal set of operations—a sensor formalized as an identity function, set-theoretic state refresh, and three fundamental forms of reference chains (reference, counter-reference, and semi-reference)—and rigorously derives core cognitive functions including state sequences, demand, comparison, efficiency, and finite-horizon probabilistic planning. Rather than modeling neural mechanisms, PST constitutes a design specification for any system that must maintain internal consistency while acting under incomplete information and irreversible risk. The framework offers novel resolutions to classical problems such as Russell's paradox, the cognitive status of Gödelian incompleteness, the grounding of negative feedback, and the comprehension of film editing. The primary purpose of this paper is to establish, through the public academic record, the originality and completeness of the Predictive Set Theory framework.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
NANQ: Noise-Floor-Aware Mixed-Precision Non-Uniform Quantization for Analog Compute-in-Memory
Yizhe Chen, Wenshuai Yao, Saiya Wang et al. (+6 more)
Analog compute-in-
memory (CIM) enables energy-efficient neural network inference, but device variation and read noise can severely degrade low-bit quantized models. Existing CIM-oriented quantization methods mainly minimize ideal quantization error, ignoring the hardware noise floor and thus causing inefficient precision allocation. We propose NANQ, a noise-aware mixed-precision non-uniform quantization framework for analog CIM. NANQ models magnitude-dependent weight noise from measured responses of an eFlash CIM array and converts the noise profile into an adaptive quantization density, assigning finer resolution to low-noise regions while avoiding ineffective precision in noise-dominated regions. It further assigns layer-wise bit-widths by identifying each layer's precision saturation point under hardware noise using a unified threshold. On-chip experiments on an eFlash CIM SoC show that, under 2-bit weight-magnitude quantization, NANQ improves
vision-model accuracy by 8.05 percentage points and reduces
language-model PPL by 54.7% on average over PowerQuant. Mixed-precision NANQ captures most of the gains obtainable from additional quantization resources with only 3.2-3.8 equivalent bits.
Read the full abstract ▾
All authors Yizhe Chen, Wenshuai Yao, Saiya Wang, Yuannuo Feng, Wenbo Qi, Kechao Tang, Ngai Wong, Wenyong Zhou, Wang Kang
Artificial Intelligence
Explainable AI for the EU Right to Explanation: A Systematic Review of the Law-XAI Translation Gap
Benjamin Fresz, Elena Dubovitskaya and Marco F. Huber
When
algorithms make or influence consequential decisions—about loan eligibility, hiring, or healthcare—EU law grants affected individuals a Right to Explanation. Yet whether (and how) Explainable AI (XAI) can satisfy this right in practice remains poorly understood, with direct implications for individuals' ability to contest automated decisions that affect their lives. This paper presents a systematic literature review of XAI in the context of the EU Right to Explanation, with particular focus on Art. 15(1)(h) GDPR, Art. 86 AI Act (AIA), and related instruments. We consider papers published from 2024 onwards, as the final version of the AIA was published in July 2024—with Art. 86 being added late. From 2643 initial records identified by a deliberately broad search, we review 57 full texts, of which only 19 papers demonstrate substantive integration of both legal and technical perspectives, showing gaps in the interdisciplinary synthesis of the current regulatory framework. We document three problematic patterns across the corpus: Most misidentify the GDPR legal basis; few engage with the CJEU's Dun & Bradstreet judgment (likely due to publication timing); and the distinction between explanation form (governed by addressee) and content (governed by legal purpose) is often conflated. We conceptualize this as the Addressee/Purpose Framework, propose a four-phase blueprint for operationalization, and identify six concrete open research questions. Without further progress, the Right to Explanation risks remaining a formal obligation without a technically realizable path to compliance.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CR
Steganalysis of Adaptive Covert Collusion in Tool-Using Agent Populations: A Black-Box, Cross-Principal Approach
Mohamed Chahine Ghanem
Tool-using
agents built on large
language models (LLMs) are increasingly deployed not by a single operator but by many, side by side on shared infrastructure. This creates a population-level risk that single-
agent safeguards miss: a handful of
agents can quietly coordinate, rigging a
market, boosting one another in a review process, or timing a joint data grab, while each one looks perfectly well-behaved. The difficulty is that the organisations running these
agents cannot see inside one another's models, so any realistic detector must work from behaviour alone: black-box, trace-only, and often with only partial visibility. We treat covert coordination as an information-hiding problem and build a black-box steganalysis detector that combines cross-run mutual-information estimation, permutation tests, distributional-shift statistics, and timing and tool-call side channels, all calibrated to a fixed false-positive budget. Our central move is to stop testing against a single fixed code: we pit the detector against an adversary that continually rewrites its encoding to slip past whatever the detector has learned, and we run this red-versus-blue contest in tool-using,
memory-carrying environments rather than toy games. Capacity theory then tells us what to expect, a detection-capacity frontier, a covert bit-rate below which black-box detection is provably no better than chance. We set out an experiment to map this frontier, report clearly labelled placeholder results pending measurement, and flag a practical evasion, spreading a payload across sessions, that current methods largely miss.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Measuring Explainer Stability via Attribution Separability
Eddie Conti, Álvaro Parafita and Axel Brando
Attribution methods (AMs) assign an importance score to each feature and are widely adopted to explain black-box models. However, most methods can produce variable attribution scores due to
stochastic components in their definition. In this paper, we propose a distribution-based framework to capture the stability of attribution scores. In particular, our approach allows to understand the degree of separability in the ranked attribution vector and obtain the largest index for which a feature ranking remains reliable. We further extend this framework to compare AMs based on the robustness of their rankings across a
dataset. Through experiments, we demonstrate how to apply our method to evaluate explainer stability. Overall, our approach provides a complementary criterion for evaluating the stability of AMs.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Output-Aware Rotation for INT2 KV-Cache Quantization
Vincent-Daniel Yun, Woosang Lim, Minsoo Cheong et al. (+4 more)
The key-value (KV) cache has become a major
memory and bandwidth bottleneck in long-context large
language model inference, making ultra-low-bit quantization increasingly important. However, existing rotation-based INT2 methods optimize cache statistics or proxy errors before the complete
attention readout, even though the model is ultimately affected by the error propagated through
attention and the output projection $W_O$. To address this mismatch, we propose OptR, an output-aware rotation method that minimizes post-$W_O$
attention-output error. OptR decomposes the post-$W_O$
attention-output error into key- and value-induced terms and learns per-head orthogonal corrections through the full INT2 quantization and
attention path. OptR further applies an
attention-equivalent key reparameterization to reduce large channel-wise offsets without changing the softmax distribution. Across three models and five
reasoning and coding
benchmarks, OptR consistently improves both QuaRot and OSCAR and strengthens long-context
retrieval, while preserving the paged KV-cache format with negligible inference overhead.
Read the full abstract ▾
All authors Vincent-Daniel Yun, Woosang Lim, Minsoo Cheong, Sunwoo Lee, Murali Annavaram, Sai Praneeth Karimireddy, Sungjoo Yoo
Artificial Intelligence · Cross-listed from cs.LG
Learning Molecular Representations from Cellular Phenotypes with Structure Preservation
Xuan Lin, Jingyu Sheng, Tengfei Ma et al. (+2 more)
Phenotypic drug discovery enables the discovery of functional relationships between molecular structures and cellular responses. However, existing
multimodal representation learning methods often optimize cross-modal
alignment without considering the intrinsic organization of chemical space, resulting in distorted molecular representations and loss of structural information. We propose PhenMol, a structure-preserving framework for phenotype-aware molecular representation learning. PhenMol disentangles molecular and cellular representations into shared and private components, enabling phenotype-guided
alignment while preserving chemical structures through a dedicated molecular branch. This design integrates cellular phenotype information without disrupting molecular neighborhood organization. Experiments on approximately $3.04 \times 10^{4}$ molecule–
cell morphology pairs demonstrate that PhenMol improves molecular property prediction across 270 bioactivity tasks, molecule–phenotype
retrieval, and clinical trial outcome prediction. Moreover, ECFP4-based structural analysis shows that PhenMol better preserves molecular neighborhoods and reduces
embedding distortion compared with existing
multimodal alignment methods. These results highlight the importance of structure-aware constraints in
multimodal molecular representation learning and provide an effective approach for integrating cellular phenotypes with chemical knowledge for drug discovery.
Read the full abstract ▾
All authors Xuan Lin, Jingyu Sheng, Tengfei Ma, Li Sun, Dapeng Xiong
Artificial Intelligence · Cross-listed from cs.SE
BulkPR-Bench: Benchmarking Queue-Level Governance of Interacting Pull Requests
Zetong Xiong, Qiao Zhao, Jun Zhang et al. (+20 more)
Coding-
agent benchmarks increasingly cover long-horizon, end-to-end, and interactive development, but typically retain one requested outcome or a fixed change sequence. Sequential policies can process a pull-request (PR) queue one candidate at a time, but when queued PRs interact, maximizing safe delivery can require jointly deciding which changes to merge and in what order. We introduce BulkPR-Bench, an executable
benchmark in which an
agent must recover consequential PR relations and return a large safe subset in executable order under a rolling-release protocol. The suite contains 581 newly authored candidate PRs on frozen snapshots of 18 real repositories. Registered state-by-state repository execution, including hidden safety checks, validates the gold relation graph; an exact oracle then computes the largest safe subset. Our primary metric, Relational Delivery Score (RDS), scores safe delivery and correct rejection over relation groups from the realized merge trace; Global Safety-Gated Yield (Global-SGY) separately measures strict delivery of the realized whole-queue plan. Under the buffered primary protocol with batch size $K=32$, the three highest RDS estimates among the six models are 66.6%, 62.0%, and 57.9%, compared with 53.1% for the strongest sequential baseline. Only 8 of 324 model runs complete a queue exactly. Critical-relation recall ranges from 35.2% to 57.7%, and diagnostic runs supplied with the gold relations show substantial remaining headroom. Gains on relation groups therefore do not yet translate into dependable whole-queue governance.
Read the full abstract ▾
All authors Zetong Xiong, Qiao Zhao, Jun Zhang, Xueying Lyu, Zhi Li, Yixiang Tu, Xiaowen Yang, Yunjie Zhang, Yufeng Wang, Zhe Zhang, Kaize Yu, Hanwen Du, Zhongkai Sun, Zhuoxin Liu, Zekun Lin, Jianwen Yang, Ruining Chen, Ying Zhang, Tingxuan Pan, Ke Chen, Shubin Han, Chuanhao Sun, Yehua Yang
Artificial Intelligence · Cross-listed from q-bio.QM
A Blind Spot in Alignment: Quantifying Biosecurity Risks in Large Language Models
Shu Quan, Tianfang Hao, Sitong Fang et al. (+8 more)
Large
Language Models (LLMs) are accelerating biological research, yet this same capability poses a critical biosecurity threat: models that assist in
protein engineering can equally be prompted to generate predicted toxin-like sequences, potentially lowering the barrier to biological misuse. Current safety evaluations, however, operate in natural
language and cannot determine whether a model-generated amino acid sequence is biological gibberish or a computational risk signal. To address this evaluation blind spot, we introduce SPIKE-Bench, coupling 631 curated toxin-design
prompts across seven functional categories with the SPIKE funnel, a three-stage protocol that filters output through compliance, biological plausibility, and predicted toxicity, producing stage-level diagnostics and an aggregate function-aware metric: the Functional Harmfulness Rate (FHR). An audit of 32 LLMs reveals that most models freely comply with toxin-design requests; FHR is driven primarily by biological generation capability rather than safety
alignment, reaching 50.7%; and Refusal Rate fails to predict functional risk. As a first step toward mitigation, we provide BioSafe-Guard, a domain-specialized classifier that substantially reduces predicted functional risk while preserving benign utility. We release SPIKE-Bench and BioSafe-Guard at https://github.com/PKU-
Alignment/SPIKE-Bench to support more rigorous biosecurity evaluation of LLMs.
Read the full abstract ▾
All authors Shu Quan, Tianfang Hao, Sitong Fang, He Geng, Jiayi Zhou, Boyuan Chen, Kaile Wang, Donghai Hong, Juntao Dai, Yaodong Yang, Jiaming Ji
Artificial Intelligence · Cross-listed from cs.CR
$S^3$: Improving Agent Safety through Multi-Stage Defense
Zibo Xiao, Haoyu Wang and Jun Sun
Large
Language Model (LLM)
agents rely on multi-stage agentic workflows, with stages such as
memory, planning, and tool execution, to accomplish complex tasks. However, risks may emerge at different stages, propagate across steps, and become difficult to detect and mitigate. Existing safety methods protect only isolated stages and are difficult to integrate, leaving
agents without comprehensive protection throughout the workflow. To address these limitations, we introduce Stage-Specific Safety Skills, a unified abstraction that represents heterogeneous safety designs as reusable and composable components with explicit stage semantics. We further develop an automated transformation pipeline that converts existing safety designs into reusable safety skills and establish a community-driven safety skill library. Building on this abstraction, we propose $S^3$, a multi-stage defense framework in which a guard
agent orchestrates stage-specific safety skills for risk detection and mitigation throughout the agentic workflow. We also construct the Multi-Stage Risk
Benchmark (MSRB) to evaluate representative risks across workflow stages. Experimental results show that $S^3$ consistently outperforms representative state-of-the-art baselines in both safety effectiveness and utility preservation. These results demonstrate the potential of stage-specific safety skills as a scalable and composable foundation for building resilient and trustworthy
agent systems.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.SE
TraceCompiler: Skill-Guided Mining and Compilation of LLM Agent Traces into Mostly Deterministic Workflows
Salma El Yadouni and Guanyi Li
Tool-using
language-model
agents repeatedly rediscover procedures they have already executed, producing traces that mix reusable structure with retries, exploration, accidental ordering, and repeated lookups. We present TraceCompiler, a skill-guided system that mines clusters of noisy
agent traces and compiles them into executable, mostly deterministic workflows. It admits an inter-tool dependency only when a consumer argument contains a value attributable uniquely to an earlier producer; every hard edge carries an auditable evidence tuple, and ambiguous relations are marked suspected and impose no ordering constraint. Bindings are classified as constants, user inputs, copied outputs, transforms, or residual LLM decisions. On T1, a mechanized form of the rule recovers producer-consumer dependencies at 0.928 precision and 0.943 recall over 15,775 def-use edges of its training split, against 0.711 F1 for adjacency and 0.712 for a frequency-thresholded directly-follows measure on identical data; the compiler skill run blind reaches 0.992 on 250 of those edges. On AppWorld we replay released trajectories in the deterministic simulator to recover masked return values and measure the rule against 563
token edges at 0.993 precision - a self-consistency check, since replay injects
tokens by a related heuristic. We compile two recurring intents: a Venmo money-request intent reduces 34 observed API calls to 11 runtime calls and, under leave-one-out execution against the
benchmark's own state tests, passes 15 of 21, the failing fold escalating rather than acting because its required branch was never observed; and a Spotify/Todoist intent the compiler correctly refuses to compile, because an irreversible side effect is under-determined. We measure call reduction but not offline compilation cost, so we claim no net efficiency result.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from eess.SY
Bridging Artificial Intelligence and Power Systems Education Using a Hands-On Executable Framework
Junjie Yin, Buxin She, Xinyu Feng et al. (+1 more)
Artificial intelligence (AI) is increasingly central to power and energy systems, supporting modeling,
forecasting, optimization, and control. Yet most existing works emphasize specialized applications and offer little reusable material for newcomers or interdisciplinary learners, who increasingly rely on large
language models rather than building their own. This gap points to a need for engineering-grounded AI (EGAI), in which AI workflows follow established engineering and power-system domain rules rather than acting as task-agnostic black boxes. Motivated by a community survey of researchers and practitioners, which shows 92% report at least one barrier before running an AI model and 94% want a power-specific hands-on course. This paper presents a framework consisting of open, executable module library that lowers the entry barrier for AI in power systems. The modules follow a progressive difficulty ladder that maps core AI concepts onto representative power-system tasks: (i) foundational deep neural network (DNN) templates for function approximation and load-curve fitting; (ii) a domain-coupled convolutional neural network (CNN) power-flow surrogate for a 5-bus system; and (iii) frontier modules on DNN-assisted optimization, deep reinforcement learning (DRL) for battery storage control, and physics-informed neural networks (PINNs) for the swing equation. All modules are released as Jupyter notebooks that run locally or on Google Colab and are delivered through an IEEE online course and IEEE Power & Energy Society (PES) webinar series. The webinar drew more than 590 live attendees, which is among the ten most-attended IEEE PES webinars, and over 344 repository visits within two weeks, reinforcing the survey-based motivation.
Read the full abstract ▾
All authors Junjie Yin, Buxin She, Xinyu Feng, Fangxing Li
Artificial Intelligence · Cross-listed from cs.CV
Tingyu Song, Mingxin Li, Yanzhao Zhang et al. (+5 more)
Sparse
retrieval underpins modern search systems, from web search to
retrieval-augmented generation. Existing work has introduced Learned Sparse
Retrieval (LSR) to push beyond exact lexical matching toward richer semantics. Yet LSR has so far remained tied to encoder-style bidirectional architectures, and its extension to
multimodal settings still relies heavily on auxiliary cross-modal modules. To address these limitations, we introduce UEmbed (Unified
Embedding), a decoder-only
multimodal embedding model that produces both sparse lexical and dense representations in one causal forward pass. UEmbed appends N learnable special
tokens to the input and partitions the vocabulary into N disjoint subsets. Each
token's causal hidden state predicts sparse weights over its assigned subset, and the N subsets are concatenated into the full sparse vector. Trained on public data, we release UEmbed at 2B, 4B, and 9B scales. UEmbed-9B reaches 71.8 (dense) and 71.0 (sparse) on MMEB-v2, outperforming
multimodal embedding models trained on publicly available data (e.g., RzenEmbed). On BEIR, UEmbed also remains competitive with strong dense and sparse baselines. Furthermore, we demonstrate the practical utility of UEmbed across three dimensions: effectiveness, efficiency, and agentic applications. Overall, UEmbed offers a new paradigm: it unifies dense and sparse
embeddings in one model, while further extending sparse
retrieval to unify text and
multimodal inputs.
Read the full abstract ▾
All authors Tingyu Song, Mingxin Li, Yanzhao Zhang, Dingkun Long, Pengjun Xie, Zhijie Nie, Yilun Zhao, Shu Wu