Papers

Test-time adaptation (TTA) can mitigate domain shift without source data, but it is highly brittle under adversarially contaminated test streams, where corrupted inputs also destabilize online updates. We study robust test-time adaptation (RTTA) in the adversarial-stream setting, which remains comparatively underexplored relative to standard TTA, and propose SAFER (Stochastic Augmentation Framework for Enhanced Robustness), a training-free reliability-guided augmentation wrapper for RTTA. SAFER preserves the wrapped TTA objective while replacing brittle single-view predictions with a reliability-guided pooled predictor. For each test sample, SAFER generates stochastic augmentations and aggregates their predictions through correlation-weighted pooling with outlier detection. We further study an adaptive-mixing extension that improves clean-performance retention by adjusting original-versus-augmentation weighting using feature disagreement signals. We evaluate on PACS, VLCS, and OfficeHome under PGD attacks at various attack rates. Across benchmarks, SAFER improves resilience of TTA methods to adversarial attacks while maintaining competitive clean performance.

Reinforcement Learning (RL) has been widely applied to sequential decision-making, yet it often suffers from poor sample efficiency due to costly interactions with the environment. A limited line of recent work has started exploring improving RL efficiency by leveraging external knowledge expressed in natural-language instructions. However, the few existing approaches typically treat the entire instruction as a single conditioning input, failing to account for the stage-dependent nature of language guidance, especially in complex environments. In this paper, we propose Hierarchical Reinforcement Learning with Language Instructions (HRLLI), a hierarchical RL framework that explicitly models natural-language instructions as dynamically selectable semantic guidance during decision-making. HRLLI decomposes instructions into a set of piecewise guidance elements, where each instruction piece may become relevant at different stages of interaction with the environment. A novel hierarchical RL policy structure is then formulated in a Select-to-Act paradigm: a high-level semantic policy acts as a guidance selector that selects the most relevant instruction piece to the current state to guide the low-level agent's decision, while a low-level policy executes environment actions conditioned on the selected guidance. The two-level policies are learned simultaneously to maximize augmented expected returns from interactions with the environment. This design enables the agent to adaptively ground language instructions into stage-specific decisions during interaction. Experiments on the instruction-intensive RTFM benchmark show that HRLLI consistently outperforms strong instruction-conditioned RL baselines, demonstrating that explicitly modeling adaptive instruction selection significantly improves the effectiveness of RL.

Principled regression for stochastic processes is a long-standing challenge with deep connections to scientific inverse problems. We introduce Flow Annealing Posterior Sampling (FLAPS), to our knowledge the first function-space posterior sampling framework that unifies stochastic-process regression and PDE inverse problems. Built on pretrained function-space flow-matching priors, FLAPS enables likelihood-guided posterior inference from sparse and noisy observations, supports variable query discretizations, and avoids explicit prior-density evaluation. Its Langevin correction uses a low-rank covariance preconditioner to exploit dominant function-space correlations across discretizations. Across Gaussian and non-Gaussian stochastic-process regression benchmarks and diverse PDE inverse problems, FLAPS produces coherent posterior samples with well-calibrated uncertainty quantification, significantly outperforming existing functional regression baselines and achieving competitive or better noisy PDE inverse performance than diffusion-based posterior samplers while reducing test-time sampling cost.

Robots deployed in realistic settings will accumulate experience across many sessions and tasks over their deployment. The robot's tasks may often require it to remember information from multiple sessions ago, making long-context robot memory important for real-world deployments. However, most robot-memory benchmarks today are based on single episodes or a short context. To measure how current robot memory systems perform on longer sessions with more distractions, we introduce RoboMME-Interference, a cross-session benchmark built on RoboMME (Dai et al., 2026). For each query episode, we construct a session history using the query's relevant prior demonstration followed by a controlled number of unrelated sessions, which we provide to the VLA as memory and measure accuracy. Running RoboMME's released memory-augmented $\pi_{0.5}$ variants unmodified through this benchmark, we find that while perceptual memory variants improve success when given the history without any distractors, they decay strongly and steadily as unrelated sessions accumulate. The subgoal variants, which read the history with a vision-language model and pass written subgoals to the policy, improve less at their best but hold more of that improvement as distractors accumulate. Adding a retrieval step to the strongest perceptual variant, which selects the section of history most visually similar to the robot's current view and passes only that section to the policy, restores its no-distractor success rate at every interference level. With this release, we emphasize the importance of long-context memory and robustness to interference and show that current systems largely fail on such capabilities. The project page, videos, code, and data are at https://robotmemorybench.com.

External skills can improve action-oriented LLM agents without changing model weights, but persistent skill updates are risky when they are distilled from sparse or noisy trajectories. A plausible reflection may encode a useful procedure, a spurious shortcut, or a rule that the target executor cannot reliably follow. We propose Hypothesis-Driven Skill Optimization (HDSO), a train-free framework in which both the skill curator and the agent executor are frozen inference endpoints. The curator observes executor traces, proposes a falsifiable hypothesis with an explicit validation plan, instantiates the hypothesis as a candidate skill package, validates the package through paired control/treatment executions, reviews behavior differences, and consolidates only supported candidates into an approved repository. The executor consumes approved skills through progressive disclosure, preserving the executor-only path when no skill is selected. On ALFWorld, HDSO improves executor-only baselines by +6.9 Avg. SR points for Qwen3-8B and +4.0 points for Qwen3.6-27B. Under 20% randomly flipped success/failure feedback during skill discovery and validation, HDSO preserves a +7.1-point gain for Qwen3-8B. Transfer and heterogeneous-pair diagnostics further show that validated repositories can be useful beyond the run that produced them, but cross-model curation succeeds only when curator diagnosis, executor capability, and validation evidence align. HDSO provides an auditable skill lifecycle for frozen action agents rather than an unconstrained memory accumulation procedure.

LLM-as-a-judge has become the dominant approach to scalable evaluation in NLP pipelines, yet judges themselves carry systematic biases that raw accuracy hides: they favor responses placed in slot A (position bias), they prefer longer responses regardless of quality (verbosity bias), and their reliability degrades sharply in lower-resource languages. We introduce BabelJudge, an open-source benchmark and reliability audit framework that measures all four failure modes – position bias, verbosity bias, order inconsistency, and cross-lingual degradation – on any judge model, without requiring human preference labels. The key insight is gold-labelling by degradation: starting from a high-quality reference response and applying a controlled perturbation yields a pairwise item whose gold label is known by construction, eliminating annotation cost. We evaluate Qwen2.5-7B-Instruct-4bit across English, Hindi, Arabic, and Swahili and find that our composite bias-penalised reliability score drops from 0.714 in Hindi to 0.550 in Swahili, a gap that raw accuracy (0.835 vs. 0.660) understates. Swahili order consistency collapses to 0.480, meaning judge verdicts are near-random under slot-order swaps – a failure mode invisible to accuracy alone. We further extend the framework to agentic evaluation via nine trajectory-level perturbations (argument corruption, tool swaps, hallucinated calls, missing steps) and three new metrics: tool accuracy, hallucination detection rate, and trajectory-length bias. BabelJudge is released as a Python package supporting 11 judge backends. Code: https://github.com/Shreyaskc/BabelJudge

The rapid growth of interactive Large Language Model serving has made efficient management of dynamic Key-Value cache footprints increasingly important for inference performance. Modern inference systems overwhelmingly rely on time-centric scheduling heuristics, such as Shortest Job First. However, their classical guarantees are rooted in traditional scheduling modeling, failing to capture the highly dynamic, 2D spatio-temporal geometric growth specific to LLM inference mechanisms. To resolve this, we propose the geometry-aware online scheduling by introducing the Smallest Volume First (SVF) algorithm and its highly efficient variant, 1-bit SVF. Via a novel volume-certificate proof, we establish a worst-case approximation ratio for SVF that approaches 3 in the high-concurrency regime of LLM serving, substantially improving upon the prior best constant of 48. Building upon this core breakthrough, we complete a comprehensive theoretical taxonomy analyzing our algorithms across different traffic scenarios and information availability. Practically, we seamlessly integrate our approach as a plug-and-play layer in vLLM. Extensive evaluations on Llama-3.1 models demonstrate comprehensive performance gains: SVF delivers strong reductions in both average and tail latency, while 1-bit SVF, with merely a single bit of information, achieves competitive throughput and latency. This work establishes a theoretically sound and empirically proven approach for resolving memory-constrained scheduling in modern LLM deployments. To facilitate future research, our code is available at https://github.com/Li-Beverly-Kong/Geometry-Aware-Online-Scheduling.git.

Attention sinks, representation collapse, and norm stratification are treated as transformer-specific pathologies. We show they are not specific to attention: they are what content-based routing does under a fixed similarity metric. We give a reframing identity: softmax attention is Boltzmann-weighted aggregation over Euclidean distances with constant key norms, so its score omits a $-\|k\|^2$ term and is blind to key magnitude. This predicts that any router whose metric is ill-matched to its representations should compensate, by concentrating its routing and collapsing the routed representations. We test it on routers that score and aggregate over different axes: softmax attention over tokens (nine pretrained transformers), graph attention over nodes, a selective state-space model and a recurrent mixer over time, and learned residuals over depth. All develop the same signature, and two within-model ablations show it is caused by the routing mechanism rather than by incidental dynamics. The form is contingent, set by the strength of the positional brake each router carries alongside its content score; we sweep that brake and move the onset across its whole range. The mechanism is not contingent, and it does not require norm stratification: a router with norm-normalized keys concentrates just the same. We do not claim these models implement Riemannian geometry; the geometric view is a diagnostic that names the inadequacy of the flat, norm-blind metric.

Reinforcement learning with verifiable rewards (RLVR) is widely viewed as a promising path toward continuously improving large language models. Recent works, however, suggest that mainstream RLVR often reallocates sampling probabilities among trajectories already present in the base model: it can improve sampling efficiency, reflected by higher pass@1 scores, but yields limited gains, and can even decrease pass@k scores when k is large, and therefore may fail to expand the base model's reasoning capacity boundary. In this paper, we present a boundary-aware Curriculum RL approach to move beyond the base model's reasoning capacity boundary. Our approach first uses pass@k sampling to locate the current reasoning capacity boundary, then applies targeted teacher guidance to examples near or beyond that boundary, and finally uses RL to consolidate the newly introduced reasoning patterns. Across Qwen, Llama, and DeepSeek base models, boundary-aware Curriculum RL improves both pass@1 scores and pass@256 scores, with pass@1 reflecting one-attempt performance and pass@256 serving as an empirical proxy for the reasoning capacity boundary. In our experiments, average pass@256 improves by 9.8 percentage points over the base models and by 10.3 percentage points over Vanilla RLVR. These results suggest that boundary-aware Curriculum RL can provide a scalable route for LLMs to continuously improve beyond the base model's empirical reasoning capacity boundary.

Path-based attribution methods such as Integrated Gradients (IG) are widely adopted for their strong axiomatic properties and effectiveness in attributing model predictions to input features by integrating gradients along a path from a baseline to the input. However, the choice of the attribution path largely affects the quality of explanations, and existing approaches rely on fixed or hand-crafted paths that often produce noisy or distorted attributions. To address this limitation, we propose Diffusion Integrated Gradients (DiffIG), a novel method that reformulates path generation as a conditional generative modeling problem. DiffIG first trains a diffusion model to learn a distribution over paths generated from a Stick-Breaking Process, then employs guided sampling to embed user guidance during the sampling procedure. We demonstrate that DiffIG quantitatively matches or outperforms existing path-based methods, achieving perceptually aligned explanations. This work introduces a new generative perspective for flexible, inference-time controllable Explainable Artificial Intelligence (XAI) methods.

Data augmentation (DA) has been proven to be an effective means for improving protein representation learning (PRL) by generating additional training samples. Although mainstream perturbation- and sampling-based augmentation methods can produce data containing sufficient variations, they carry the risk of disrupting the protein structure and function. Some crafted protein homology modeling tools can generate conformations, but reduce structural diversity. The above dilemmas lead us to a question: Can we restore the disrupted structure caused by DA operations, providing data with both the original structure and diverse variations? In this work, we first analyze and empirically reveal the structure defect and performance degradation issues of existing DA methods. Based on the findings, we propose a simple yet effective DA method, Manifold Restore Mixing (MRM), for protein representation learning. Specifically, inspired by manifold mixup, we mix the hidden representations of original and augmented protein data to generate new samples that restore structural information lost in DA while introducing diverse variations. Furthermore, we develop a sample difficulty scheduler that adjusts the beta distribution in mixup to provide models with progressively challenging mixed samples during training, which improves the final performance. Comprehensive experiments on various PRL backbones and downstream tasks demonstrate the effectiveness and generalization of our method. The complete code and weights will be released upon acceptance. We provide a implementation at https://github.com/KingGugu/MRM.

In the rapidly evolving field of software development, code stylometry analyzing unique stylistic signatures of programmers plays a crit-ical role in authorship attribution and cybersecurity. Recent advancements in artificial intelligence, particularly Large Language Models (LLMs) like GPT-3.5 and GPT-4, have introduced new dimensions to this field, challenging traditional stylometry techniques. This study investigates the effectiveness of LLMs in altering code stylometry while preserving functionality and evaluates the impact of various prompt engineering strategies. Through comprehensive experiments, we assess how well these models can obscure stylistic signatures to avoid detection by a Random Forest classifier trained for authorship attribution. The results reveal significant differences in effectiveness between single-shot and multi-shot methods and highlight the importance of detailed, structured prompts. Additionally, functionality preservation checks demonstrate the challenges in maintaining code integrity post-modification. This research provides critical insights into the robustness of authorship attribution techniques against advanced AI capabilities, informing future cybersecurity and software engineering developments

Edge Internet of Things (IoT) agents are often constrained by memory capacity, privacy requirements, communication latency, and recurring inference cost. Current smart-home assistants commonly rely on API-level command interfaces or cloud-based language models that remain difficult to deploy on edge devices. This paper addresses edge IoT command generation as a many-to-one structured output task, where multiple natural-language instructions map to the same canonical command string for deterministic smart-home parsing. To support this setting, we propose Semantic-Conditioned Edge-Aware Neural Framework for Structured IoT Command Generation (SCENIC), an end-to-end framework covering model architecture selection, Smart Home Instruct data generation, triplet-loss contrastive supervised fine-tuning, pruning and quantization, and deployment-oriented export. We evaluate sub-0.2B-scale transformer backbones, which are, to the best of our knowledge, among the smallest language-model backbones studied for edge IoT structured command generation. On Smart Home Instruct-Bench, the strongest dense decoder-only row reaches 99.0% EM@1, while the encoder-decoder model retains stronger high-sparsity behavior. A representative pruned INT8 encoder-decoder export preserves 91.0% EM@1 and 99.0% EM@5 while reducing exported model size by 25.38%. TensorRT profiling of the NVIDIA 2:4 sparse encoder export further shows up to 1.8x encoder-component speedup, indicating that the selected encoder-decoder deployment path can retain structured command accuracy under edge-oriented compression while hardware acceleration evidence remains component-level. The SCENIC code and experimental artifacts are open sourced to support reproducibility.

We study trajectory optimization in mobile sensing systems that must identify which member of a finite candidate set is the true target, while maintaining reachability to all potential candidate targets, under resource constraints. Deferred-Decision Trajectory Optimization (DDTO) addresses this setting by computing trajectories that reach individual targets but remain coincident for as long as possible before separating toward different targets. We propose Active-Sensing DDTO (AS-DDTO), which extends DDTO by adding a trajectory-dependent information-acquisition term to the planning objective. The resulting planner maintains reachability to candidate targets while biasing the coincident portion of the trajectories toward regions that enable earlier target identification. The framework supports Bayesian updates and conformal candidate-set updates for distance-dependent sensing. We derive a mixed-integer conic reformulation and provide guarantees on recursive feasibility, belief concentration, and fixed-time coverage for the raw conformal candidate set. Numerical simulations show improved target identification compared with standard DDTO under distance-dependent sensing uncertainty and limited sensing budget.

In competitive first-person shooter (FPS) games such as Counter-Strike 2 (CS2), account-integrity review often asks whether an account's recent behavior remains consistent with its historical operator. This consistency question arises in cases such as temporary substitution, rank boosting, and high-skill players using lower-ranked accounts, where manual review requires comparing a current match against multiple historical matches. We formulate this review task as same-player verification: we encode the behavioral trajectory of a single player in a match replay (demo) as a demo-player behavioral fingerprint, and train a model to judge whether two behavioral observations come from the same real player. Using CS2-specific domain knowledge, the fingerprints cover crosshair control, movement-stop-fire coordination, economy/buy, combat/engagement, and temporal rhythm. We construct strict six-fold evaluations on the Perfect dataset (3,570 demos and 35,700 demo-player observations) and the Professional dataset (539 demos and 5,390 demo-player observations). The final pairwise model reaches ROC AUCs of 0.926 and 0.956, respectively. Feature analysis shows that the strongest identity signals come from aiming/crosshair and other low-level mechanical behaviors, indicating that stable mechanics are more informative for this verification task than single-match performance outcomes. On fixed eligible query cohorts, aggregating pairwise evidence between a current demo and multiple historical demos raises account-history AUC on Perfect from 0.923 at K=1 to 0.982 at K=10, and on Professional from 0.914 at K=1 to 0.975 at K=5. These results show that CS2 demo behavior can support supervised same-player verification and account-level identity-consistency modeling through multi-demo history aggregation.

Agentic AI is gaining new insights and advancements in the field of Artificial Intelligence, fostering significant potential to enable rapid transformation across various domains.This rapid advancement and the potential to revolutionize various domains advocate the need for a deeper understanding and firm grasp of the technology. Moreover, an investigation into state of the art research directions in agentic AI needs to be conducted to comprehensively assess the potential scope for improvement and application.Therefore, to address these objectives, a comprehensive review can provide researchers and practitioners with valuable insights into the current state and future research scopes of agentic AI.Hence, this work considers the recently published scholarly contributions in agentic AI across various domains and discusses the fundamentals and working principles of Agentic AI, traces the historical and theoretical evolution of agency in artificial systems, explores and discusses Agentic AIs architecture, working principles, and functionalities, explores real-world applications of Agentic AI across various domains, analyzes the research findings, identifies current challenges, and discuss potential future research directions, and proposes a comprehensive framework of stakeholders intention to use and adopt Agentic AI with the help of proposed system quality dimensions.Therefore, this systematic review provides researchers and practitioners with a comprehensive understanding of Agentic AI, its current developments and applications, highlights key research gaps, and outlines future research directions.

Research on human-GenAI collaboration yields conflicting findings: GenAI can enhance creativity yet reduce collective diversity, with uneven benefits across skill levels. Rather than treating these as contradictions, we argue they reflect a core feature of GenAI: abundance. GenAI makes ideas, drafts, and recombinations plentiful, potentially expanding the hypothesis space and surfacing unanticipated possibilities. However, abundance alone doesn't ensure better outcomes. We propose generative fit as a unifying mechanism explaining when abundance yields productive creativity and when it backfires. Drawing on Generativity Theory, generative fit captures how well a system's generative potential complements a community's generative capacities. We develop a conceptual framework for collaborative human-GenAI settings where participants share goals, depend on one another, and must integrate diverse contributions. By mapping abundance to cognitive, social, and organizational factors of collective creativity, we explain apparent tradeoffs and offer actionable implications for designing workflows that convert abundance into valued creative outcomes.

Diffusion Large Language Models (dLLMs) have emerged as a promising alternative to autoregressive (AR) LLMs, generating tokens in parallel. This makes them effective draft models for speculative decoding (SD), producing an entire block of draft tokens in a single forward pass. Yet existing diffusion-based drafting methods rely on linear drafting, even though dLLMs emit multiple candidate tokens across positions, inducing a large combinatorial space of decoding paths. Consequently, they limit acceptance length and decoding efficiency. To exploit this multi-candidate structure, we apply tree-based drafting to diffusion drafters, enabling exploration of diverse candidate paths. However, we find that naive tree drafting is suboptimal: diffusion marginals are prefix-blind, mismatching the prefix-based AR verification and yielding unreliable path ranking. We propose PRESTO, a principled framework that extends tree-based drafting to diffusion drafters while resolving the fundamental mismatch between diffusion draft confidence and prefix-based AR verification through PREfix-aligned Scoring and priority-based Tree search for diffusion speculative decOding. The key principles behind PRESTO are that (1) candidate ranking should align with the prefix-based nature of AR verification, and (2) tree construction should prioritize candidate paths with high verification potential to maximize acceptance length. Extensive experiments show that PRESTO achieves up to an average of $1.5\times$ end-to-end throughput speedup on the state-of-the-art dedicated diffusion drafter SD and an average of $1.12\times$ on self-speculative diffusion LLMs across diverse benchmarks.

Table Question Answering (TableQA) aims to reason over tables to answer user queries. Existing research treats all questions uniformly and evaluates solely through overall accuracy, obscuring a critical reality that LLMs excel at simple lookups yet struggle with complex operations like aggregation and arithmetic. To reveal this disparity, we introduce a novel Operation-wise TableQA task with a fine-grained question taxonomy and release two datasets named WikiTQ-ow and TabFact-ow for evaluation. As for modeling bottlenecks, existing methods flatten tables into linearized texts, disrupting inherent structures and inducing the “lost-in-the-middle” issue, which poses a primary barrier to complex cross-row reasoning. Moreover, they typically reason from scratch, neglecting reusable patterns shared across similar operations. To address these limitations, we propose a Skill-augmented Table Graph Reasoning (SkillTGR) framework for self-evolving structured reasoning. Specifically, SkillTGR represents tables as attributed graphs with explicit row-column-cell structures, where LLMs plan and execute dynamic chains to retrieve evidence subgraphs for graph traversal reasoning. Based on this, SkillTGR builds a hierarchical SkillBank to distill reason trajectories into abstract skills under cognitive heuristics, then hybrid retrieves both successful and failed skills for contrastive augmented table graph reasoning, thereby enabling the continual self-evolution. Extensive experiments demonstrate that SkillTGR achieves superior performance with an average of 5.91% overall and 5.06% operation-wise improvement, also reducing 19.76% token consumption and 27.64% inference latency. Our codes and data will be released upon publication.

Despite rapid progress in LLM-based code generation, writing correct and performant kernels for hardware accelerators remains a key bottleneck in scaling modern ML workloads. We present MKEvolve (Modular Kernel Evolve), a framework that iteratively co-evolves a modular decomposition of complex PyTorch modules and the LLM-generated kernel for each submodule, refining the decomposition by splitting and fusing across iterations while independently improving each subkernel via LLM-driven beam search. The resulting kernels are programmatic compositions of independently verified subkernels, making them configurable (subkernel implementations are swappable), interpretable (errors and speedups are traceable to specific subkernels), and readily adaptable to related model architectures. Experiments with Triton on KernelBench L2 and L3, spanning multi-operator sequences and full model architectures, show that MKEvolve improves both correctness and speedup over end-to-end direct synthesis baselines while reducing LLM token usage by up to 35%.

Large Language Models (LLMs) perform strongly on well-specified reasoning tasks with a feasible answer. However, problems encountered in the open world can become ill-posed due to inconsistent conditions, conflicting statements, or mutually incompatible requirements, admitting no valid responses. We argue that reasoning of such ill-posed problems involving conflicts require novel LLM capabilities to make hidden conflicts explicit, maintain competing hypotheses via multiple reasoning branches, and generate alternative responses in a single pass, all of which are challenging due to the limitation of the next-token prediction mechanism in LLMs. To this end, we propose FlowEdit, a novel framework that leverages information-theoretic principles to quantify and regulate internal reasoning flows of LLMs, for generating a full set of alternative responses under valid hypotheses. FlowEdit can be viewed as enforcing a branch-aware reasoning process using two dual information-theoretic objectives on the model's internal reasoning representations: maximizing the information flow from each selected hypothesis to the branch outcome, while minimizing the overlap and conditional dependence across sibling branches, to provide a diverse, informative set of responses with broad coverage. We show that this is achieved through tractable variational bounds under boundary embeddings being {\epsilon}-sufficient, optimizing the underlying conditional mutual information in LLM reasoning process. Extensive experiments demonstrate that FlowEdit outperforms leading proprietary models, improving exact-set-match accuracy by 68%, while boosting overall response informativeness by 24%. We further show that flow regulation surfaces in the token stream as a redistribution of next-token entropy that concentrates inside each branch, amplifies at flow boundaries, and scales with the number of flows the problem requires.

Recent latent reasoning methods, such as CODI and COCONUT, face a fundamental interpretability problem: they maintain multiple superimposed candidate traces in the hidden space at each step, unlike explicit- CoT, which follows a single transparent reasoning trace. Existing mechanistic methods show compression, shortcuts, and superposition without explaining how reasoning evolves across latent steps. To address this gap, we model latent token sequences as trajectories in representation space and apply dynamical systems analysis to characterize the evolution of reasoning. Using quantitative measures, such as step-to-step change, direction consistency, and Lyapunov sensitivity, alongside qualitative projections, such as UMAP and DMD/PHATE, we show that latent CoT exhibits structured, non-random dynamics with two distinct stability classes. CODI behaves as a stable attractor, while COCONUT behaves as an unstable expanding system, and SIM-CoT supervision tightens both behaviors without changing the underlying dynamics. This framework advances the interpretability of latent CoT reasoning dynamics and provides actionable insights for improving latent reasoning performance. Code1 and Project page2 available online.

Deep neural models for click-through rate prediction often exhibit a sharp decline in validation performance immediately after the first training epoch despite continued improvement in training loss. This instability restricts effective learning and limits model performance. In this study, we analyze this behavior using large-scale industrial datasets and evaluate practical mitigation strategies. While reducing the learning rate provides only incremental gains, controlling feature sparsity yields substantial improvements. Removing highly sparse features and aggregating infrequent feature values stabilizes training, extends useful learning beyond a single epoch, and improves both offline evaluation metrics and online system performance.

Symbolic regression via genetic programming routinely fails on small, wide datasets - a regime common in clinical-trial monitoring, biostatistics, and engineering pilot studies - by converging on bloated, overfit expressions that exploit correlation rather than prediction. We present Evolutional Math, an open-source genetic programming system that combines four design choices to yield compact, interpretable formulas in this regime. First, fitness is measured by R-squared on held-out cross-validation folds rather than Pearson correlation on the training set, eliminating single-variable shortcuts that correlate but mis-scale. Second, a multi-island architecture runs independent populations seeded with distinct operator subsets (algebraic, logarithmic, trigonometric, and full) with ring-topology migration every M generations, preventing the search from collapsing into one region of formula space. Third, a structural deduplication scheme treats formulas differing only in constants as equivalent, so the elite archive contains structurally distinct candidates rather than near-duplicate variants. Fourth, top-k individuals undergo numerical constant refinement via scipy L-BFGS-B after each migration phase, decoupling structure search from parameter fitting. We evaluate the system on synthetic benchmarks of the form log(x_i) * x_j / (x_k * c), trigonometric mixtures, and an anonymized clinical site-monitoring dataset with 24 rows and approximately 290 candidate numeric features. The system consistently recovers compact ground-truth structures with R-squared at or above 0.99 within tens of thousands of unique formula evaluations. A reference implementation is released under a noncommercial source-available license.

The intricate structures of biological neural networks largely emerge during development, guided by a comparatively compressed blueprint encoded in the genome. The connectivity that emerges from this decoding process is rich in structure, and already equips the organism with functional modules upon birth. This initial structure serves as a scaffold that can be gradually refined and fine-tuned through lifelong experience, via a variety of plasticity mechanisms. Drawing inspiration from this interaction between evolutionary and developmental modes of learning, we use hypernetworks to learn a compressed generative process that generates the connectivity of a modular reservoir. We show that this marriage between curriculum-based meta-learning and modular reservoir computing can generate sparse recurrent networks that solve difficult temporal tasks with minimal training and without concessions to robustness.

Low-resource African languages lack text corpora needed for language model training. We investigate whether ASR pipelines can extend text resources for two typologically distinct West African languages: Fongbe (tonal, diacritic-rich) and Hausa (non-tonal). We fine-tune MMS-300M on a curated 12.3-hour Fongbe dataset, achieving 9.48% WER on the ALFFA benchmark - a 78% relative reduction from the prior 44.04% baseline - while preserving tonal diacritics critical to the language. For Hausa, we apply an existing fine-tuned Whisper-Small model. We catalog 1,553 YouTube videos (236 hours) and process a subset of 424 videos (45.49 hours) selected to balance domain diversity with available computational resources, producing 6,770 transcribed segments. Human evaluation on 50 randomly sampled segments per language shows mean quality scores of 57.4/100 for Hausa and 36.5/100 for Fongbe, indicating that while Hausa transcriptions approach acceptable quality for corpus construction, Fongbe transcriptions require post-processing or improved models for production use. We release the curated dataset, fine-tuned model, transcribed corpus, and full video catalog following platform terms and ethical guidelines.

Applying pre-trained language models to new domains through full fine-tuning is computationally expensive and prone to catastrophic forgetting. To address this limitation, we introduce a novel parameter-efficient strategy for unsupervised domain adaptation that combines a custom PEFT architecture with mixed-objective training. The proposed method integrates invertible adapters with Low-Rank Adaptation (LoRA) and jointly optimizes classification on labeled source-domain data and masked language modeling on unlabeled target-domain data. This joint training scheme supports task adaptation while preserving knowledge of the target domain. We evaluate the method on the Multi-Genre Natural Language Inference (MNLI) dataset across 20 domain shifts. Our approach achieves average performance improvements of 1.41 percentage points over the parameter-efficient state-of-the-art UDapter, 1.26 percentage points over the fully tuned DANN baseline, and 0.86 percentage points over DSN, while updating only 7% of the model parameters. These findings establish a new state-of-the-art result for parameter-efficient unsupervised domain adaptation and demonstrate that carefully designed PEFT combinations with concurrent optimization can outperform both parameter-efficient and conventional fully tuned approaches.

We investigate the translation quality of current large language models (LLMs) for English-to-Hausa and English-to-Fongbe - two typologically distinct West African languages from the Afroasiatic and Niger-Congo families respectively - and evaluate whether standard automatic metrics reliably reflect human judgment for these low-resource languages. We evaluate four models (GPT-4o Mini, Claude Sonnet 4, Gemini 2.5 Flash, and Qwen2.5-7B) at progressive scales (500 to 10,000 sentences) using automatic metrics (BLEU, chrF++, TER, COMET, BERTScore) validated against native-speaker judgment. Our results reveal three key findings. First, translation quality varies substantially by language: Hausa achieves acceptable quality (human scores 4.0-4.5/5) while Fongbe achieves poor quality (1.0-2.2/5), with a consistent 3x BLEU gap across all systems. Second, model rankings differ by language - Gemini leads for Fongbe while GPT-4o leads for Hausa by human evaluation - indicating that performance on one low-resource African language does not predict performance on another. Third, metric-human correlation varies dramatically: perfect rank correlation for Fongbe (rho=1.0) but weak correlation for Hausa (rho=0.5), where human evaluators preferred GPT-4o despite all automatic metrics ranking Claude first. We further show that neural metrics like BERTScore exhibit embedding collapse (within-language similarity >0.99) for both languages, limiting their ability to differentiate translation quality. Based on these findings, we recommend multi-metric evaluation for low-resource African languages, with particular caution when interpreting neural metrics. We establish that minimum sample sizes of n=2,500 sentences are required for stable system rankings, as smaller samples produced artifact findings that reversed at scale.

Memory safety vulnerabilities remain a significant threat even for projects with extensive fuzzing and manual auditing. Recent results suggest that large language models hold great promise for detecting such vulnerabilities, but they are unreliable, at risk of hallucination, and challenging to scale to repository-size codebases. This paper presents Revelio, a cost-efficient end-to-end agentic framework for memory-safety vulnerability discovery. Revelio addresses the problem of hallucination by generating an executable Proof-of-Vulnerability, which is checked with a deterministic sanitizer. It reduces cost using inexpensive LLMs and lightweight static analysis to help generate and rank vulnerability hypotheses, reporting vulnerabilities only when they can be reproduced and confirmed by a sanitizer. We evaluated Revelio on seven production-quality projects that had been continuously fuzzed for five to eight years, as well as on 100 randomly selected Arvo projects from the CyberGym benchmark. With around one hour per project and a total cost of $300, Revelio discovered 19 previously unknown memory-safety vulnerabilities. On benchmarks, Revelio outperformed frontier coding agents across diverse backbone models at comparable token costs. Our results suggest that Revelio enables scalable and trustworthy end-to-end LLM-based memory-safety vulnerability detection.

Electroencephalogram (EEG) analysis remains the clinical gold standard for epilepsy diagnosis and seizure detection. While Deep Learning (DL) has significantly advanced automated EEG interpretation, its transition from controlled experimental settings to routine clinical deployment is severely bottlenecked by fundamental architectural flaws. Standard DL models operate as opaque black-boxes lacking clinical interpretability, demand massive amounts of balanced annotated data, and incur steep computational costs incompatible with resource-constrained wearable or implantable neuromodulation devices. This paper presents a comprehensive review of these prevailing limitations and explores Kolmogorov-Arnold Networks (KANs) as a emerging paradigm for EEG-based seizure detection. By replacing the fixed activation functions of traditional neurons with flexible, learnable functions along the network's connections, KANs bridge the critical gap between predictive accuracy and mathematical transparency. We systematically analyze how KAN architectures resolve the shortcomings of traditional DL-based models by offering exceptional parameter efficiency, inherent interpretability for physician trust, and robust performance under data scarcity. Ultimately, this review establishes KANs not merely as an incremental algorithmic update, but as a fundamental paradigm shift necessary to actualize next-generation, patient-specific, and thoroughly transparent clinical EEG monitoring systems.