Subword tokenization is a key design choice for modern language models, including large language models (LLMs), with byte- and character-level BPE serving as a widely used baseline. Standard BPE selects merges by raw pair frequency, which favors compression but can conflate true adjacency cohesion with pairs that are frequent due to high marginal counts. This paper introduces Significance-Gain BPE, a drop-in alternative merge criterion that measures cohesion via a z-statistic under an independence null model and combines it with an explicit compression-aware gain term. Significance-Gain BPE is evaluated on WikiText-103 (raw) character slices using a small causal Transformer language model, reporting both token-dependent perplexity and the tokenizer-invariant metric bits per character (BPC). At a representative operating point, Significance-Gain BPE reduces validation and test perplexity by 13% and 12%, respectively, and improves validation and test BPC by about 0.9 to 1.0%. A vocabulary-size sweep further shows lower BPC in most closest-compression comparisons, suggesting that statistically grounded merge selection can improve predictive efficiency per unit of raw text across a range of compression regimes.

Sign Language Machine Translation (SLMT) aims to bridge communication between Deaf and hearing individuals. However, its progress is constrained by scarce datasets, limited signer diversity, and large domain gaps between sign motion patterns and pretrained representations. Existing transfer learning approaches in SLMT are static and often lead to overfitting. These challenges call for the development of an adaptive framework that preserves pretrained structure while remaining robust across linguistic and signing variations. To fill this void, we propose a Hierarchical Adaptive Transfer Learning (HATL) framework, where pretrained layers are progressively and dynamically unfrozen based on training performance behavior. HATL combines dynamic unfreezing, layer-wise learning rate decay, and stability mechanisms to preserve generic representations while adapting to sign characteristics. We evaluate HATL on Sign2Text and Sign2Gloss2Text translation tasks using a pretrained ST-GCN++ backbone for feature extraction and the Transformer and an adaptive transformer (ADAT)for translation. To ensure robust multilingual generalization, we evaluate the proposed approach across three datasets: RWTH-PHOENIXWeather-2014 (PHOENIX14T), Isharah, and MedASL. Experimental results show that HATL consistently outperforms traditional transfer learning approaches across tasks and models, with ADAT achieving BLEU-4 improvements of 15.0% on PHOENIX14T and Isharah and 37.6% on MedASL.

Taiwanese Hokkien (Taigi) presents unique opportunities for advancing speech technology methodologies that can generalize to diverse linguistic contexts. We introduce Breeze Taigi, a comprehensive framework centered on standardized benchmarks for evaluating Taigi speech recognition and synthesis systems. Our primary contribution is a reproducible evaluation methodology that leverages parallel Taiwanese Mandarin resources. We provide 30 carefully curated Mandarin-Taigi audio pairs from Taiwan's Executive Yuan public service announcements with normalized ground truth transcriptions. We establish Character Error Rate (CER) as the standard metric and implement normalization procedures to enable fair cross-system comparisons. To demonstrate the benchmark's utility and provide reference implementations, we develop speech recognition and synthesis models through a methodology that leverages existing Taiwanese Mandarin resources and large-scale synthetic data generation. In particular, we fine-tune a Whisper model on approximately 10,000 hours of Taigi synthetic speech data. Our ASR model achieves 30.13% average CER on the benchmark, outperforming existing commercial and research systems. By providing standardized evaluation protocols, diverse training datasets, and open baseline models, we offer a replicable framework with methodologies applicable to various linguistic contexts.

While differentially private (DP) fine-tuning of large language models (LLMs) is a powerful tool, it is often computationally prohibitive or infeasible when state-of-the-art models are only accessible via proprietary APIs. In such settings, generating DP synthetic data has emerged as a crucial alternative, offering the added benefits of arbitrary reuse across downstream tasks and transparent exploratory data analysis without the opaque constraints of a model's parameter space. Private Evolution (PE) is a promising API-based framework for this goal; however, its performance critically depends on initialization. When the private data distribution deviates substantially from the foundation model's pre-training priors--particularly in highly specialized domains--PE frequently struggles to align with the target data, resulting in degraded utility, poor convergence, and inefficient API usage. To address this initialization bottleneck, we propose Metadata Augmented Private Language Evolution (MAPLE). MAPLE leverages differentially private tabular metadata extraction and in-context learning to effectively ground the initial synthetic distribution in the target domain. Extensive experiments on challenging, domain-specific text generation tasks demonstrate that MAPLE achieves a significantly more favorable privacy-utility trade-off, converges faster, and drastically reduces API costs compared to previous PE methods.

Real-world path planning tasks typically involve multiple constraints beyond simple route optimization, such as the number of routes, maximum route length, depot locations, and task-specific requirements. Traditional approaches rely on dedicated formulations and algorithms for each problem variant, making them difficult to scale across diverse scenarios. In this work, we propose a flexible framework that leverages large language models (LLMs) to solve constrained path planning problems directly from natural language input. The core idea is to allow users to describe routing tasks conversationally, while enabling the LLM to interpret and solve the problem through solution verification and iterative refinement. The proposed method consists of two integrated components. For problem types that have been previously formulated and studied, the LLM first matches the input request to a known problem formulation in a library of pre-defined templates. For novel or unseen problem instances, the LLM autonomously infers a problem representation from the natural language description and constructs a suitable formulation in an in-context learning manner. In both cases, an iterative solution generation and verification process guides the LLM toward producing feasible and increasingly optimal solutions. Candidate solutions are compared and refined through multiple rounds of self-correction, inspired by genetic-algorithm-style refinement. We present the design, implementation, and evaluation of this LLM-based framework, demonstrating its capability to handle a variety of constrained path planning problems. This method provides a scalable and generalizable approach for solving real-world routing tasks with minimal human intervention, while enabling flexible problem specification through natural language.

Bengali is spoken by over 230 million people yet remains severely under-served in automatic speech recognition (ASR) and speaker diarization research. In this paper, we present our system for the DL Sprint 4.0 Bengali Long-Form Speech Recognition (Task~1) and Bengali Speaker Diarization Challenge (Task~2). For Task~1, we propose a data-centric pipeline that constructs a high-quality training corpus from Bengali YouTube audiobooks and dramas \cite{tabib2026bengaliloop}, incorporating LLM-assisted language normalization, fuzzy-matching-based chunk boundary validation, and muffled-zone augmentation. Fine-tuning the \texttt{tugstugi/whisper-medium} model on approximately 21,000 data points with beam size 5, we achieve a Word Error Rate (WER) of 16.751 on the public leaderboard and 15.551 on the private test set. For Task~2, we fine-tune the pyannote.audio community-1 segmentation model with targeted hyperparameter optimization under an extreme low-resource setting (10 training files), achieving a Diarization Error Rate (DER) of 0.19974 on the public leaderboard, and .26723 on the private test set. Our results demonstrate that careful data engineering and domain-adaptive fine-tuning can yield competitive performance for Bengali speech processing even without large annotated corpora.

Despite the strong performance of Large Language Models (LLMs) on complex instruction-following tasks, precise control of output length remains a persistent challenge. Existing methods primarily attempt to enforce length constraints by externally imposing length signals or optimization objectives, while largely overlooking the underlying limitation: the model's intrinsic deficit in length cognition. To address this, we propose LARFT (Length-Aware Reinforcement Fine-Tuning), a training framework that aligns the model's length cognition with its action. Specifically, LARFT integrates length-oriented reinforcement learning with a hindsight length awareness. By transforming on-policy data into hindsight self-awareness tasks where the model learns to identify the actual length of its own generation, LARFT jointly optimizes the model's internal representation of length information and refines its policy to satisfy length constraints, thereby achieving precise and reliable length instruction following. Extensive experiments across four base models demonstrate that LARFT outperforms existing baselines, achieving an average improvement of +20.92 points across three length instruction following benchmarks with only a marginal decline of -1.45 points on four general capability benchmarks.

Argument mining (AM) is an interdisciplinary research field focused on the automatic identification and classification of argumentative components, such as claims and premises, and the relationships between them. Recent advances in large language models (LLMs) have significantly improved the performance of argument classification compared to traditional machine learning approaches. This study presents a comprehensive evaluation of several state-of-the-art LLMs, including GPT-5.2, Llama 4, and DeepSeek, on large publicly available argument classification corpora such as Args.me and UKP. The evaluation incorporates advanced prompting strategies, including Chain-of- Thought prompting, prompt rephrasing, voting, and certainty-based classification. Both quantitative performance metrics and qualitative error analysis are conducted to assess model behavior. The best-performing model in the study (GPT-5.2) achieves a classification accuracy of 78.0% (UKP) and 91.9% (Args.me). The use of prompt rephrasing, multi-prompt voting, and certainty estimation further improves classification performance and robustness. These techniques increase the accuracy and F1 metric of the models by typically a few percentage points (from 2% to 8%). However, qualitative analysis reveals systematic failure modes shared across models, including instabilities with respect to prompt formulation, difficulties in detecting implicit criticism, interpreting complex argument structures, and aligning arguments with specific claims. This work contributes the first comprehensive evaluation that combines quantitative benchmarking and qualitative error analysis on multiple argument mining datasets using advanced LLM prompting strategies.

Evaluating the symbolic reasoning of large language models (LLMs) calls for geometry benchmarks that require multi-step proofs grounded in both text and diagrams. However, existing benchmarks are often limited in scale and rarely provide visually grounded multiple-choice questions, limiting reliable evaluation of complex reasoning. We introduce GeoChallenge, a dataset of 90K automatically generated multiple-choice geometry proof problems, each requiring multi-step reasoning over aligned textual descriptions and diagrams. GeoChallenge provides fine-grained complexity ratings and formal language annotations to enable controlled evaluation. Experiments on multiple advanced LLMs show a clear performance gap between models and humans (the best-performing model, GPT-5-nano, achieves 75.89 exact match vs. 94.74 for humans). Further analysis also reveals three common failure patterns of LLMs: (1) exact match failures under the multiple-choice setting; (2) weak visual reliance; and (3) overextended reasoning without convergence.

Large Language Models (LLMs) perform well in short contexts but degrade on long legal documents, often producing hallucinations such as incorrect clauses or precedents. In the legal domain, where precision is critical, such errors undermine reliability and trust. Retrieval Augmented Generation (RAG) helps ground outputs but remains limited in legal settings, especially with small, locally deployed models required for data privacy. We identify two failure modes: retrieval errors due to lexical redundancy in legal corpora, and decoding errors where models generate answers despite insufficient context. To address this, we propose Metadata Enriched Hybrid RAG to improve document level retrieval, and apply Direct Preference Optimization (DPO) to enforce safe refusal when context is inadequate. Together, these methods improve grounding, reliability, and safety in legal language models.

Healthcare question-answering (QA) systems face a persistent challenge: users submit queries with spelling errors at rates substantially higher than those found in the professional documents they search. This paper presents the first controlled study of spelling correction as a retrieval preprocessing step in healthcare QA using real consumer queries. We conduct an error census across two public datasets -- the TREC 2017 LiveQA Medical track (104 consumer health questions) and HealthSearchQA (4,436 health queries from Google autocomplete) -- finding that 61.5% of real medical queries contain at least one spelling error, with a token-level error rate of 11.0%. We evaluate four correction methods -- conservative edit distance, standard edit distance (Levenshtein), context-aware candidate ranking, and SymSpell -- across three experimental conditions: uncorrected queries against an uncorrected corpus (baseline), uncorrected queries against a corrected corpus, and fully corrected queries against a corrected corpus. Using BM25 and TF-IDF cosine retrieval over 1,935 MedQuAD answer passages with TREC relevance judgments, we find that query correction substantially improves retrieval -- edit distance and context-aware correction achieve MRR improvements of +9.2% and NDCG@10 improvements of +8.3% over the uncorrected baseline. Critically, correcting only the corpus without correcting queries yields minimal improvement (+0.5% MRR), confirming that query-side correction is the key intervention. We complement these results with a 100-sample error analysis categorising correction outcomes per method and provide evidence-based recommendations for practitioners.

Immersive conversational systems in production face a persistent trade-off between responsiveness and long-horizon task capability. Real-time interaction is achievable for lightweight turns, but requests involving planning and tool invocation (e.g., search and media generation) produce heavy-tail execution latency that degrades turn-taking, persona consistency, and user trust. To address this challenge, we propose DuCCAE (Conversation while Collaboration with Augmentation and Evolution), a hybrid engine for immersive conversation deployed within Baidu Search, serving millions of users. DuCCAE decouples real-time response generation from asynchronous agentic execution and synchronizes them via a shared state that maintains session context and execution traces, enabling asynchronous results to be integrated back into the ongoing dialogue. The system orchestrates five subsystems-Info, Conversation, Collaboration, Augmentation, and Evolution-to support multi-agent collaboration and continuous improvement. We evaluate DuCCAE through a comprehensive framework that combines offline benchmarking on the Du-Interact dataset and large-scale production evaluation within Baidu Search. Experimental results demonstrate that DuCCAE outperforms strong baselines in agentic execution reliability and dialogue quality while reducing latency to fit strict real-time budgets. Crucially, deployment metrics since June 2025 confirm substantial real-world effectiveness, evidenced by a tripling of Day-7 user retention to 34.2% and a surge in the complex task completion rate to 65.2%. Our hybrid architecture successfully preserves conversational continuity while enabling reliable agentic execution, offering practical guidelines for deploying scalable agentic systems in industrial settings.

Large Language Models (LLMs) are increasingly integrated into high-stakes applications, making robust safety guarantees a central practical and commercial concern. Existing safety evaluations predominantly rely on fixed collections of harmful prompts, implicitly assuming non-adaptive adversaries and thereby overlooking realistic attack scenarios in which inputs are iteratively refined to evade safeguards. In this work, we examine the vulnerability of contemporary language models to automated, adversarial prompt refinement. We repurpose black-box prompt optimization techniques, originally designed to improve performance on benign tasks, to systematically search for safety failures. Using DSPy, we apply three such optimizers to prompts drawn from HarmfulQA and JailbreakBench, explicitly optimizing toward a continuous danger score in the range 0 to 1 provided by an independent evaluator model (GPT-5.1). Our results demonstrate a substantial reduction in effective safety safeguards, with the effects being especially pronounced for open-source small language models. For example, the average danger score of Qwen 3 8B increases from 0.09 in its baseline setting to 0.79 after optimization. These findings suggest that static benchmarks may underestimate residual risk, indicating that automated, adaptive red-teaming is a necessary component of robust safety evaluation.

Ambiguity poses a major challenge to large language models (LLMs) used as robotic planners. In this letter, we present Scene Graph-Chain-of-Thought (SG-CoT), a two-stage framework where LLMs iteratively query a scene graph representation of the environment to detect and clarify ambiguities. First, a structured scene graph representation of the environment is constructed from input observations, capturing objects, their attributes, and relationships with other objects. Second, the LLM is equipped with retrieval functions to query portions of the scene graph that are relevant to the provided instruction. This grounds the reasoning process of the LLM in the observation, increasing the reliability of robotic planners under ambiguous situations. SG-CoT also allows the LLM to identify the source of ambiguity and pose a relevant disambiguation question to the user or another robot. Extensive experimentation demonstrates that SG-CoT consistently outperforms prior methods, with a minimum of 10% improvement in question accuracy and a minimum success rate increase of 4% in single-agent and 15% in multi-agent environments, validating its effectiveness for more generalizable robot planning.

A central challenge in image-based Model-Based Reinforcement Learning (MBRL) is to learn representations that distill essential information from irrelevant visual details. While promising, reconstruction-based methods often waste capacity on large task-irrelevant regions. Decoder-free methods instead learn robust representations by leveraging Data Augmentation (DA), but reliance on such external regularizers limits versatility. We propose R2-Dreamer, a decoder-free MBRL framework with a self-supervised objective that serves as an internal regularizer, preventing representation collapse without resorting to DA. The core of our method is a redundancy-reduction objective inspired by Barlow Twins, which can be easily integrated into existing frameworks. On DeepMind Control Suite and Meta-World, R2-Dreamer is competitive with strong baselines such as DreamerV3 and TD-MPC2 while training 1.59x faster than DreamerV3, and yields substantial gains on DMC-Subtle with tiny task-relevant objects. These results suggest that an effective internal regularizer can enable versatile, high-performance decoder-free MBRL. Code is available at https://github.com/NM512/r2dreamer.

Skeleton-based action recognition is crucial for multimedia applications but heavily relies on power-hungry Artificial Neural Networks (ANNs), limiting their deployment on resource-constrained edge devices. Spiking Neural Networks (SNNs) provide an energy-efficient alternative; however, existing spiking models for skeleton data often compromise the intrinsic sparsity of SNNs by resorting to dense matrix aggregations, heavy multimodal fusion modules, or non-sparse frequency domain transformations. Furthermore, they severely suffer from the short-term amnesia of spiking neurons. In this paper, we propose the Spiking State-Space Topology Transformer (S3T-Former), which, to the best of our knowledge, is the first purely spike-driven Transformer architecture specifically designed for energy-efficient skeleton action recognition. Rather than relying on heavy fusion overhead, we formulate a Multi-Stream Anatomical Spiking Embedding (M-ASE) that acts as a generalized kinematic differential operator, elegantly transforming multimodal skeleton features into heterogeneous, highly sparse event streams. To achieve true topological and temporal sparsity, we introduce Lateral Spiking Topology Routing (LSTR) for on-demand conditional spike propagation, and a Spiking State-Space (S3) Engine to systematically capture long-range temporal dynamics without non-sparse spectral workarounds. Extensive experiments on multiple large-scale datasets demonstrate that S3T-Former achieves highly competitive accuracy while theoretically reducing energy consumption compared to classic ANNs, establishing a new state-of-the-art for energy-efficient neuromorphic action recognition.

Recent Audio Multimodal Large Language Models (Audio MLLMs) demonstrate impressive performance on speech benchmarks, yet it remains unclear whether these models genuinely process acoustic signals or rely on text-based semantic inference. To systematically study this question, we introduce DEAF (Diagnostic Evaluation of Acoustic Faithfulness), a benchmark of over 2,700 conflict stimuli spanning three acoustic dimensions: emotional prosody, background sounds, and speaker identity. Then, we design a controlled multi-level evaluation framework that progressively increases textual influence, ranging from semantic conflicts in the content to misleading prompts and their combination, allowing us to disentangle content-driven bias from prompt-induced sycophancy. We further introduce diagnostic metrics to quantify model reliance on textual cues over acoustic signals. Our evaluation of seven Audio MLLMs reveals a consistent pattern of text dominance: models are sensitive to acoustic variations, yet predictions are predominantly driven by textual inputs, revealing a gap between high performance on standard speech benchmarks and genuine acoustic understanding.

Socio-environmental planning under deep uncertainty requires researchers to identify and conceptualize problems before exploring policies and deploying plans. In practice and model-based planning approaches, this problem conceptualization process often relies on participatory modeling to translate stakeholders' natural-language descriptions into a quantitative model, making this process complex and time-consuming. To facilitate this process, we propose a templated workflow that uses large language models for an initial conceptualization process. During the workflow, researchers can use large language models to identify the essential model components from stakeholders' intuitive problem descriptions, explore their diverse perspectives approaching the problem, assemble these components into a unified model, and eventually implement the model in Python through iterative communication. These results will facilitate the subsequent socio-environmental planning under deep uncertainty steps. Using ChatGPT 5.2 Instant, we demonstrated this workflow on the lake problem and an electricity market problem, both of which demonstrate socio-environmental planning problems. In both cases, acceptable outputs were obtained after a few iterations with human verification and refinement. These experiments indicated that large language models can serve as an effective tool for facilitating participatory modeling in the problem conceptualization process in socio-environmental planning.

Aspect-Based Sentiment Intensity Analysis (ABSIA) has garnered increasing attention, though research largely focuses on domain-specific, sentence-level settings. In contrast, document-level ABSIA--particularly in addressing complex tasks like extracting Aspect-Category-Opinion-Sentiment-Intensity (ACOSI) tuples--remains underexplored. In this work, we introduce DanceHA, a multi-agent framework designed for open-ended, document-level ABSIA with informal writing styles. DanceHA has two main components: Dance, which employs a divide-and-conquer strategy to decompose the long-context ABSIA task into smaller, manageable sub-tasks for collaboration among specialized agents; and HA, Human-AI collaboration for annotation. We release Inf-ABSIA, a multi-domain document-level ABSIA dataset featuring fine-grained and high-accuracy labels from DanceHA. Extensive experiments demonstrate the effectiveness of our agentic framework and show that the multi-agent knowledge in DanceHA can be effectively transferred into student models. Our results highlight the importance of the overlooked informal styles in ABSIA, as they often intensify opinions tied to specific aspects.

This paper presents a multi-agent perception-action exploration alliance, dubbed A4VL, for efficient long-video reasoning. A4VL operates in a multi-round perception-action exploration loop with a selection of VLM agents. In each round, the team of agents performs video question-answer (VideoQA) via perception exploration followed by action exploration. During perception exploration, each agent learns to extract query-specific perception clue(s) from a few sampled frames and performs clue-based alignment to find the video block(s) that are most relevant to the query-specific event. During action exploration, A4VL performs video reasoning in three steps: (1) each agent produces its initial answer with rational, (2) all agents collaboratively scores one another through cross-reviews and relevance ranking, and (3) based on whether a satisfactory consensus is reached, the decision is made either to start a new round of perception-action deliberation by pruning (e.g., filtering out the lowest performing agent) and re-staging (e.g., new-clue and matching block based perception-action exploration), or to conclude by producing its final answer. The integration of the multi-agent alliance through multi-round perception-action exploration, coupled with event-driven partitioning and cue-guided block alignment, enables A4VL to effectively scale to real world long videos while preserving high quality video reasoning. Evaluation Results on five popular VideoQA benchmarks show that A4VL outperforms 18 existing representative VLMs and 11 recent methods optimized for long-video reasoning, while achieving significantly lower inference latency. Our code is released at https://github.com/git-disl/A4VL.

Real-world robots often operate in settings where objective priorities depend on the underlying context of operation. When the underlying context is unknown apriori, multiple robots may have to coordinate to gather informative observations to infer the context, since acting based on an incorrect context can lead to misaligned and unsafe behavior. Once the underlying true context is inferred, the robots optimize their task-specific objectives in the preference order induced by the context. We formalize this problem as a Multi-Robot Context-Uncertain Stochastic Shortest Path (MR-CUSSP), which captures context-relevant information at landmark states through joint observations. Our two-stage solution approach is composed of: (1) CIMOP (Coordinated Inference for Multi-Objective Planning) to compute plans that guide robots toward informative landmarks to efficiently infer the true context, and (2) LCBS (Lexicographic Conflict-Based Search) for collision-free multi-robot path planning with lexicographic objective preferences, induced by the context. We evaluate the algorithms using three simulated domains and demonstrate its practical applicability using five mobile robots in the salp domain setup.

Remote sensing images captured from aerial perspectives often exhibit significant scale variations and complex backgrounds, posing challenges for salient object detection (SOD). Existing methods typically extract multi-level features at a single scale using uniform attention mechanisms, leading to suboptimal representations and incomplete detection results. To address these issues, we propose a GeoGran-Aware Hierarchical Feature Fusion Network (G2HFNet) that fully exploits geometric and granular cues in optical remote sensing images. Specifically, G2HFNet adopts Swin Transformer as the backbone to extract multi-level features and integrates three key modules: the multi-scale detail enhancement (MDE) module to handle object scale variations and enrich fine details, the dual-branch geo-gran complementary (DGC) module to jointly capture fine-grained details and positional information in mid-level features, and the deep semantic perception (DSP) module to refine high-level positional cues via self-attention. Additionally, a local-global guidance fusion (LGF) module is introduced to replace traditional convolutions for effective multi-level feature integration. Extensive experiments demonstrate that G2HFNet achieves high-quality saliency maps and significantly improves detection performance in challenging remote sensing scenarios.

State estimation for hybrid systems that undergo intermittent contact with their environments, such as extraplanetary robots and satellites undergoing docking operations, is difficult due to the discrete uncertainty propagation during contact. To handle this propagation, this paper presents an optimal-control-based particle filtering method that leverages saltation matrices to map out uncertainty propagation during contact events. By exploiting a path integral filtering framework that exploits the duality between smoothing and optimal control, the resulting state estimation algorithm is robust to outlier effects, flexible to non-Gaussian noise distributions, and handles challenging contact dynamics in hybrid systems. To evaluate the validity and consistency of the proposed approach, this paper tests it against strong baselines on the stochastic dynamics generated by a bouncing ball and spring loaded inverted pendulum.

Generalized planning studies the construction of solution strategies that generalize across families of planning problems sharing a common domain model, formally defined by a transition function $γ: S \times A \rightarrow S$. Classical approaches achieve such generalization through symbolic abstractions and explicit reasoning over $γ$. In contrast, recent Transformer-based planners, such as PlanGPT and Plansformer, largely cast generalized planning as direct action-sequence prediction, bypassing explicit transition modeling. While effective on in-distribution instances, these approaches typically require large datasets and model sizes, and often suffer from state drift in long-horizon settings due to the absence of explicit world-state evolution. In this work, we formulate generalized planning as a transition-model learning problem, in which a neural model explicitly approximates the successor-state function $\hatγ \approx γ$ and generates plans by rolling out symbolic state trajectories. Instead of predicting actions directly, the model autoregressively predicts intermediate world states, thereby learning the domain dynamics as an implicit world model. To study size-invariant generalization and sample efficiency, we systematically evaluate multiple state representations and neural architectures, including relational graph encodings. Our results show that learning explicit transition models yields higher out-of-distribution satisficing-plan success than direct action-sequence prediction in multiple domains, while achieving these gains with significantly fewer training instances and smaller models. This is an extended version of a short paper accepted at ICAPS 2026 under the same title.

Reinforcement learning (RL) agents under partial observability often condition actions on internally accumulated information such as memory or inferred latent context. We formalise such information-conditioned interaction patterns as behavioural dependency: variation in action selection with respect to internal information under fixed observations. This induces a probe-relative notion of $ε$-behavioural equivalence and a within-policy behavioural distance that quantifies probe sensitivity. We establish three structural results. First, the set of policies exhibiting non-trivial behavioural dependency is not closed under convex aggregation. Second, behavioural distance contracts under convex combination. Third, we prove a sufficient local condition under which gradient ascent on a skewed mixture objective decreases behavioural distance when a dominant-mode gradient aligns with the direction of steepest contraction. Minimal bandit and partially observable gridworld experiments provide controlled witnesses of these mechanisms. In the examined settings, behavioural distance decreases under convex aggregation and under continued optimisation with skewed latent priors, and in these experiments it precedes degradation under latent prior shift. These results identify structural conditions under which probe-conditioned behavioural separation is not preserved under common policy transformations.

We present Federated Learning Playground, an interactive browser-based platform inspired by and extends TensorFlow Playground that teaches core Federated Learning (FL) concepts. Users can experiment with heterogeneous client data distributions, model hyperparameters, and aggregation algorithms directly in the browser without coding or system setup, and observe their effects on client and global models through real-time visualizations, gaining intuition for challenges such as non-IID data, local overfitting, and scalability. The playground serves as an easy to use educational tool, lowering the entry barrier for newcomers to distributed AI while also offering a sandbox for rapidly prototyping and comparing FL methods. By democratizing exploration of FL, it promotes broader understanding and adoption of this important paradigm.

AI-text detectors face a critical robustness challenge: adversarial paraphrasing attacks that preserve semantics while evading detection. We introduce StealthRL, a reinforcement learning framework that stress-tests detector robustness under realistic adversarial conditions. StealthRL trains a paraphrase policy against a multi-detector ensemble using Group Relative Policy Optimization (GRPO) with LoRA adapters on Qwen3-4B, optimizing a composite reward that balances detector evasion with semantic preservation. We evaluate six attack settings (M0-M5) on the full filtered MAGE test pool (15,310 human / 14,656 AI) against four detectors: RoBERTa, Fast-DetectGPT, Binoculars, and MAGE. StealthRL achieves near-zero detection on three of the four detectors and a 0.024 mean TPR@1%FPR, reducing mean AUROC from 0.79 to 0.43 and attaining a 97.6% attack success rate. Critically, attacks transfer to two held-out detectors not seen during training, revealing shared architectural vulnerabilities rather than detector-specific brittleness. We additionally conduct LLM-based quality evaluation via Likert scoring on 500 matched samples per method, analyze detector score distributions to explain why evasion succeeds, and provide per-detector AUROC with bootstrap confidence intervals. Our results expose significant robustness gaps in current AI-text detection and establish StealthRL as a principled adversarial evaluation protocol. Code and evaluation pipeline are publicly available at https://github.com/suraj-ranganath/StealthRL.

Real-world deployment of adaptive traffic signal control, to date, remains limited due to the uncertainty associated with vision-based perception, implicit safety, and non-interpretable control policies learned and validated mainly in simulation. In this paper, we introduce UCATSC, a model-based traffic signal control system that models traffic signal control at an intersection using a stochastic decision process with constraints and under partial observability, taking into account the uncertainty associated with vision-based perception. Unlike reinforcement learning methods that learn to predict safety using reward shaping, UCATSC predicts and enforces hard constraints related to safety and starvation prevention during counterfactual rollouts in belief space. The system is designed to improve traffic delay and emission while preventing safety-critical errors and providing interpretable control policy outputs based on explicit models.

We study positional encodings for multi-view transformers that process tokens from a set of posed input images, and seek a mechanism that encodes patches uniquely, allows $SE(3)$-invariant attention with multi-frequency similarity, and can adapt to the geometry of the underlying 3D scene. We find that prior (absolute or relative) encoding schemes for multi-view attention do not meet these desiderata, and present RayRoPE to address this gap. RayRoPE represents patch positions based on associated rays and computes query-frame projective coordinates to ensure $SE(3)$ invariance. To adapt to scene geometry, RayRoPE predicts (without direct supervision) a per-token depth to obtain its position along the corresponding ray, while also modeling uncertainty and analytically computing the expected positional encoding. We validate our method on the tasks of novel-view synthesis and stereo depth estimation. While remaining efficient, RayRoPE consistently improves over alternate position encoding schemes (e.g., 24% relative improvement on LPIPS in RE10K and 15% in CO3D).

Vision-Language Navigation (VLN) aims to enable agents to navigate to a target location based on language instructions. Traditional VLN often follows a close-set assumption, i.e., training and test data share the same style of the input images and instructions. However, the real world is open and filled with various unseen environments, posing enormous difficulties for close-set methods. To this end, we focus on the General Scene Adaptation (GSA-VLN) task, aiming to learn generalized navigation ability by introducing diverse environments and inconsistent instructions.Recent research indicates that by means of fast and slow cognition systems, human beings could generate stable policies, which strengthen their adaptation for open world. Inspired by this idea, we propose the slow4fast-VLN, establishing a dynamic interactive fast-slow reasoning framework. The fast-reasoning module, an end-to-end strategy network, outputs actions via real-time input. It accumulates execution records in a history repository to build memory. The slow-reasoning module analyze the memories generated by the fast-reasoning module. Through deep reflection, it extracts experiences that enhance the generalization ability of decision-making. These experiences are structurally stored and used to continuously optimize the fast-reasoning module. Unlike traditional methods that treat fast-slow reasoning as independent mechanisms, our framework enables fast-slow interaction. By leveraging the experiences from slow reasoning, it continually improves the accuracy and generalization ability of fast decisions. This interaction allows the system to continuously adapt and efficiently execute navigation tasks when facing unseen scenarios. Extensive experiments demonstrate the superiorities of our method.