We introduce SAGE Celer 2.6, the latest in our line of general-purpose Celer models from SAGEA. Celer 2.6 is available in 5B, 10B, and 27B parameter sizes and benefits from extensive architectural modifications and further pre-training on an undisclosed model. Using our Inverse Reasoning (IR) pipeline, SAGEA natively trains Celer 2.6 to validate its own logic paths, minimizing cascading error and hallucination in complex reasoning tasks. Celer 2.6 also boasts natively integrated multimodal functionality with an end-to-end vision encoder to avoid common pitfalls in adapter-based approaches. Celer 2.6 provides highly competitive results on mathematics, coding, and general intelligence benchmarks (ACUMEN), along with low latency. Most importantly, Celer 2.6 is specifically optimized for South Asian language support, with a custom tokenizer for the Devanagari script and strong performance in both Nepali and Hindi without sacrificing English reasoning ability.

Rhetoric recognition is a critical component in automated essay scoring. By identifying rhetorical elements in student writing, AI systems can better assess linguistic and higher-order thinking skills, making it an essential task in the area of AI for education. In this paper, we leverage Large Language Models (LLMs) for the Chinese rhetoric recognition task. Specifically, we explore Low-Rank Adaptation (LoRA) based fine-tuning and in-context learning to integrate rhetoric knowledge into LLMs. We formulate the outputs as JSON to obtain structural outputs and translate keys to Chinese. To further enhance the performance, we also investigate several model ensemble methods. Our method achieves the best performance on all three tracks of CCL 2025 Chinese essay rhetoric recognition evaluation task, winning the first prize.

Mapping Cyber Threat Intelligence (CTI) text to MITRE ATT\&CK technique IDs is a critical task for understanding adversary behaviors and automating threat defense. While recent Retrieval-Augmented Generation (RAG) approaches have demonstrated promising capabilities in this domain, they fundamentally rely on a flat retrieval paradigm. By treating all techniques uniformly, these methods overlook the inherent taxonomy of the ATT\&CK framework, where techniques are structurally organized under high-level tactics. In this paper, we propose H-TechniqueRAG, a novel hierarchical RAG framework that injects this tactic-technique taxonomy as a strong inductive bias to achieve highly efficient and accurate annotation. Our approach introduces a two-stage hierarchical retrieval mechanism: it first identifies the macro-level tactics (the adversary's technical goals) and subsequently narrows the search to techniques within those tactics, effectively reducing the candidate search space by 77.5\%. To further bridge the gap between retrieval and generation, we design a tactic-aware reranking module and a hierarchy-constrained context organization strategy that mitigates LLM context overload and improves reasoning precision. Comprehensive experiments across three diverse CTI datasets demonstrate that H-TechniqueRAG not only outperforms the state-of-the-art TechniqueRAG by 3.8\% in F1 score, but also achieves a 62.4\% reduction in inference latency and a 60\% decrease in LLM API calls. Further analysis reveals that our hierarchical structural priors equip the model with superior cross-domain generalization and provide security analysts with highly interpretable, step-by-step decision paths.

Maritime distress communications transmitted over very high frequency (VHF) radio are safety-critical voice messages used to report emergencies at sea. Under the Global Maritime Distress and Safety System (GMDSS), such messages follow standardized procedures and are expected to convey essential details, including vessel identity, position, nature of the distress, and required assistance. In practice, however, automatic analysis remains difficult because distress messages are often brief, noisy, and produced under stress, may deviate from the prescribed format, and are further degraded by automatic speech recognition (ASR) errors caused by channel noise and speaker stress. This paper presents SeaAlert, an LLM-based framework for robust analysis of maritime distress communications. To address the scarcity of labeled real-world data, we develop a synthetic data generation pipeline in which an LLM produces realistic and diverse maritime messages, including challenging variants in which standard distress codewords are omitted or replaced with less explicit expressions. The generated utterances are synthesized into speech, degraded with simulated VHF noise, and transcribed by an ASR system to obtain realistic noisy transcripts.

Authors often struggle to interpret peer review feedback, deriving false hope from polite comments or feeling confused by specific low scores. To investigate this, we construct a dataset of over 30,000 ICLR 2021-2025 submissions and compare acceptance prediction performance using numerical scores versus text reviews. Our experiments reveal a significant performance gap: score-based models achieve 91% accuracy, while text-based models reach only 81% even with large language models, indicating that textual information is considerably less reliable. To explain this phenomenon, we first analyze the 9% of samples that score-based models fail to predict, finding their score distributions exhibit high kurtosis and negative skewness, which suggests that individual low scores play a decisive role in rejection even when the average score falls near the borderline. We then examine why text-based accuracy significantly lags behind scores from a review sentiment perspective, revealing the prevalence of the Politeness Principle: reviews of rejected papers still contain more positive than negative sentiment words, masking the true rejection signal and making it difficult for authors to judge outcomes from text alone.

Reliable evaluation is essential in machine learning research, yet methodological flaws-particularly data leakage-continue to undermine the validity of reported results. In this work, we investigate whether large language models (LLMs) can act as independent analytical agents capable of identifying such issues in published studies. As a case study, we analyze a gesture-recognition paper reporting near-perfect accuracy on a small, human-centered dataset. We first show that the evaluation protocol is consistent with subject-level data leakage due to non-independent training and test splits. We then assess whether this flaw can be detected independently by six state-of-the-art LLMs, each analyzing the original paper without prior context using an identical prompt. All models consistently identify the evaluation as flawed and attribute the reported performance to non-independent data partitioning, supported by indicators such as overlapping learning curves, minimal generalization gap, and near-perfect classification results. These findings suggest that LLMs can detect common methodological issues based solely on published artifacts. While not definitive, their consistent agreement highlights their potential as complementary tools for improving reproducibility and supporting scientific auditing.

The rapid digitization of nuclear power plant main control rooms has fundamentally reshaped operator interaction patterns, introducing complex soft-control behaviors and elevated cognitive risks that are not adequately addressed by existing human reliability analysis approaches. Although recent advances in large language models and autonomous agents offer new opportunities for intelligent decision support, their deployment in safety critical environments remains constrained by risks of hallucinated reasoning and weakened human authority. This study proposes NuHF Claw, a persistent cognitive-risk agent framework that enables risk governed human centered autonomy for digital nuclear operations. The core methodological innovation lies in the introduction of a risk constrained agent runtime, which tightly couples cognitive state inference with probabilistic safety assessment to regulate autonomous system behavior in real time. By integrating cognitively grounded workload and situational awareness estimation with dynamic human error probability prediction, the framework transforms conventional offline reliability analysis into a proactive intervention mechanism embedded directly within operational workflows. Experimental validation on a high-fidelity digital control room simulator demonstrates that NuHF Claw can anticipate interface induced cognitive degradation, dynamically constrain unsafe autonomous recommendations, and provide risk-aware navigational guidance while preserving human decision authority. The results highlight a fundamental shift from automation-driven operation toward cognition-aware autonomy, offering a principled pathway for the safe integration of intelligent agents into next-generation nuclear control environments.

Mobile input method editors (IMEs) are the primary interface for text input, yet they remain constrained to manual typing and struggle to produce personalized text. While lightweight large language models (LLMs) make on-device auxiliary generation feasible, enabling deeply personalized, privacy-preserving, and real-time generative IMEs poses fundamental challenges.To this end, we present HUOZIIME, a personalized on-device IME powered by LLM. We endow HUOZIIME with initial human-like prediction ability by post-training a base LLM on synthesized personalization data. Notably, a hierarchical memory mechanism is designed to continually capture and leverage user-specific input history. Furthermore, we perform systemic optimizations tailored to on-device LLMbased IME deployment, ensuring efficient and responsive operation under mobile constraints.Experiments demonstrate efficient on-device execution and high-fidelity memory-driven personalization. Code and package are available at https://github.com/Shan-HIT/HuoziIME.

Current evaluations of long-term memory in LLMs are fundamentally static. By fixating on simple retrieval and short-context inference, they neglect the multifaceted nature of complex memory systems, such as dynamic state tracking and hierarchical reasoning in continuous interactions. To overcome these limitations, we propose MemGround, a rigorous long-term memory benchmark natively grounded in rich, gamified interactive scenarios. To systematically assess these capabilities, MemGround introduces a three-tier hierarchical framework that evaluates Surface State Memory, Temporal Associative Memory, and Reasoning-Based Memory through specialized interactive tasks. Furthermore, to comprehensively quantify both memory utilization and behavioral trajectories, we propose a multi-dimensional metric suite comprising Question-Answer Score (QA Overall), Memory Fragments Unlocked (MFU), Memory Fragments with Correct Order (MFCO), and Exploration Trajectory Diagrams (ETD). Extensive experiments reveal that state-of-the-art LLMs and memory agents still struggle with sustained dynamic tracking, temporal event association, and complex reasoning derived from long-term accumulated evidence in interactive environments.

Large language models deliver strong generative performance but at the cost of massive parameter counts, memory use, and decoding latency. Prior work has shown that pruning and structured sparsity can preserve accuracy under substantial compression, while prompt-compression methods reduce latency by removing redundant input tokens. However, these two directions remain largely separate. Most model-compression methods are static and optimized offline, and they do not exploit the fact that different prompts and decoding steps activate different latent computational pathways. Prompt-compression methods reduce sequence length, but they do not adapt the executed model subnetwork. We propose a unified compressed-sensing-guided framework for dynamic LLM execution. Random measurement operators probe latent model usage, sparse recovery estimates task-conditioned and token-adaptive support sets, and the recovered supports are compiled into hardware-efficient sparse execution paths over blocks, attention heads, channels, and feed-forward substructures. The framework introduces five key contributions: task-conditioned measurements, so different prompts induce different sparse supports; token-adaptive recovery, so active substructures are re-estimated during decoding; formal sample-complexity bounds under restricted isometry or mutual incoherence assumptions; compile-to-hardware constraints that restrict recovery to GPU-efficient structures; and a joint objective that unifies prompt compression with model reduction. Together, these components recast LLM inference as a measurement-and-recovery problem with explicit approximation guarantees and deployment-oriented speedup constraints.

Ambient AI "scribe" systems promise to reduce clinical documentation burden, but automatic speech recognition (ASR) errors can remain unnoticed without careful review, and high-quality human reference transcripts are often unavailable for calibrating uncertainty. We investigate whether cross-model disagreement among heterogeneous ASR systems can act as a reference-free uncertainty signal to prioritize human verification in medical transcription workflows. Using 50 publicly available medical education audio clips (8 h 14 min), we transcribed each clip with eight ASR systems spanning commercial APIs and open-source engines. We aligned multi-model outputs, built consensus pseudo-references, and quantified token-level agreement using a majority-strength metric; we further characterized disagreements by type (content vs. punctuation/formatting) and assessed per-model agreement via leave-one-model-out (jackknife) consensus scoring. Inter-model reliability was low (ICC[2,1] = 0.131), indicating heterogeneous failure modes across systems. Across 76,398 evaluated token positions, 72.1% showed near-unanimous agreement (7-8 models), while 2.5% fell into high-risk bands (0-3 models), with high-risk mass varying from 0.7% to 11.4% across accent groups. Low-agreement regions were enriched for content disagreements, with the content fraction increasing from 53.9% to 73.9% across quintiles of high-risk mass. These results suggest that cross-model disagreement provides a sparse, localizable signal that can surface potentially unreliable transcript spans without human-verified references, enabling targeted review; clinical accuracy of flagged regions remains to be established.

Daily scenarios are characterized by visual richness, requiring Multimodal Large Language Models (MLLMs) to filter noise and identify decisive visual clues for accurate reasoning. Yet, current benchmarks predominantly aim at evaluating MLLMs' pre-existing knowledge or perceptual understanding, often neglecting the critical capability of reasoning. To bridge this gap, we introduce DailyClue, a benchmark designed for visual clue-driven reasoning in daily scenarios. Our construction is guided by two core principles: (1) strict grounding in authentic daily activities, and (2) challenging query design that necessitates more than surface-level perception. Instead of simple recognition, our questions compel MLLMs to actively explore suitable visual clues and leverage them for subsequent reasoning. To this end, we curate a comprehensive dataset spanning four major daily domains and 16 distinct subtasks. Comprehensive evaluation across MLLMs and agentic models underscores the formidable challenge posed by our benchmark. Our analysis reveals several critical insights, emphasizing that the accurate identification of visual clues is essential for robust reasoning.

Driver drowsiness significantly impairs the ability to accurately judge safe braking distances and is estimated to contribute to 10%-20% of road accidents in Europe. Traditional driver-assistance systems lack adaptability to real-time physiological states such as drowsiness. This paper proposes a deep reinforcement learning-based autonomous braking system that integrates vehicle dynamics with driver physiological data. Drowsiness is detected from ECG signals using a Recurrent Neural Network (RNN), selected through an extensive benchmark analysis of 2-minute windows with varying segmentation and overlap configurations. The inferred drowsiness state is incorporated into the observable state space of a Double-Dueling Deep Q-Network (DQN) agent, where driver impairment is modeled as an action delay. The system is implemented and evaluated in a high-fidelity CARLA simulation environment. Experimental results show that the proposed agent achieves a 99.99% success rate in avoiding collisions under both drowsy and non-drowsy conditions. These findings demonstrate the effectiveness of physiology-aware control strategies for enhancing adaptive and intelligent driving safety systems.

This paper develops a unified Markov Decision Process (MDP) framework for optimising two recurring in-match decisions in T20 cricket, namely batting order selection and bowling plan assignment, directly in terms of win and defend probability rather than expected runs. A three-phase player profile engine (Powerplay, Middle, Death) with James-Stein shrinkage (a technique that blends a player's individual statistics toward the league average when their phase-specific data is sparse) is estimated from 1,161 IPL ball-by-ball records (2008-2025). Win/defend probabilities are evaluated using vectorised Monte Carlo simulation over N = 50,000 innings trajectories. Batting orders are evaluated by comparing all feasible arrangements of the remaining players and selecting the one that maximises win probability. Bowling plans are optimised through a guided search over possible over assignments, progressively improving the allocation while respecting constraints such as the prohibition on consecutive overs by the same bowler. Applied to two 2026 IPL matches, the optimal batting order improves Mumbai Indians' win probability by 4.1 percentage points (52.4% to 56.5%), and the optimal Gujarat Titans bowling plan improves defend probability by 5.2 percentage points (39.1% to 44.3%). In both cases, the observed sub-optimality is consistent with phase-agnostic deployment: decisions that appear reasonable under aggregate metrics are shown to be costly when phase-specific profiles are applied.

Reward models are central to aligning large language models, yet they often overfit to spurious cues such as response length and overly agreeable tone. Most prior work weakens these cues directly by penalizing or controlling specific artifacts, but it does not explicitly encourage the model to ground preferences in the prompt's intent. We learn a decoder that maps a candidate answer to the latent intent embedding of the input. The reconstruction error is used as a signal to regularize the reward model training. We provide theoretical evidence that this signal emphasizes prompt-dependent information while suppressing prompt-independent shortcuts. Across math, helpfulness, and safety benchmarks, the decoder selects shorter and less sycophantic candidates with 0.877 accuracy. Incorporating this signal into RM training in Gemma-2-2B-it and Gemma-2-9B-it increases RewardBench accuracy from 0.832 to 0.868. For Best-of-N selection, our framework increases length-controlled win rates while producing shorter outputs, and remains robust to lengthening and mild off-topic drift in controlled rewrite tests.

Reasoning capabilities in large language models (LLMs) have generally advanced significantly. However, it is still challenging for existing reasoning-based LLMs to perform effective decision-making abilities in multi-agent environments, due to the absence of explicit foresight modeling. To this end, strategic reasoning, the most fundamental capability to anticipate the counterpart's behaviors and foresee its possible future actions, has been introduced to alleviate the above issues. Strategic reasoning is fundamental to effective decision-making in multi-agent environments, yet existing reasoning enhancement methods for LLMs do not explicitly capture its foresight nature. In this work, we introduce Foresight Policy Optimization (FoPO) to enhance strategic reasoning in LLMs, which integrates opponent modeling principles into policy optimization, thereby enabling explicit consideration of both self-interest and counterpart influence. Specifically, we construct two curated datasets, namely Cooperative RSA and Competitive Taboo, equipped with well-designed rules and moderate difficulty to facilitate a systematic investigation of FoPO in a self-play framework. Our experiments demonstrate that FoPO significantly enhances strategic reasoning across LLMs of varying sizes and origins. Moreover, models trained with FoPO exhibit strong generalization to out-of-domain strategic scenarios, substantially outperforming standard LLM reasoning optimization baselines.

Existing Vision Mamba-based RGB-Event(RGBE) tracking methods suffer from using static state transition matrices, which fail to adapt to variations in event sparsity. This rigidity leads to imbalanced modeling-underfitting sparse event streams and overfitting dense ones-thus degrading cross-modal fusion robustness. To address these limitations, we propose MambaTrack, a multimodal and efficient tracking framework built upon a Dynamic State Space Model(DSSM). Our contributions are twofold. First, we introduce an event-adaptive state transition mechanism that dynamically modulates the state transition matrix based on event stream density. A learnable scalar governs the state evolution rate, enabling differentiated modeling of sparse and dense event flows. Second, we develop a Gated Projection Fusion(GPF) module for robust cross-modal integration. This module projects RGB features into the event feature space and generates adaptive gates from event density and RGB confidence scores. These gates precisely control the fusion intensity, suppressing noise while preserving complementary information. Experiments show that MambaTrack achieves state-of-the-art performance on the FE108 and FELT datasets. Its lightweight design suggests potential for real-time embedded deployment.

In advanced manufacturing systems, humans and robots collaborate to conduct the production process. Effective task planning and allocation (TPA) is crucial for achieving high production efficiency, yet it remains challenging in complex and dynamic manufacturing environments. The dynamic nature of humans and robots, particularly the need to consider spatial information (e.g., humans' real-time position and the distance they need to move to complete a task), substantially complicates TPA. To address the above challenges, we decompose production tasks into manageable subtasks. We then implement a real-time hierarchical human-robot TPA algorithm, including a high-level agent for task planning and a low-level agent for task allocation. For the high-level agent, we propose an efficient buffer-based deep Q-learning method (EBQ), which reduces training time and enhances performance in production problems with long-term and sparse reward challenges. For the low-level agent, a path planning-based spatially aware method (SAP) is designed to allocate tasks to the appropriate human-robot resources, thereby achieving the corresponding sequential subtasks. We conducted experiments on a complex real-time production process in a 3D simulator. The results demonstrate that our proposed EBQ&SAP method effectively addresses human-robot TPA problems in complex and dynamic production processes.

Human-robot collaborative manufacturing, a core aspect of Industry 5.0, emphasizes ergonomics to enhance worker well-being. This paper addresses the dynamic human-robot task planning and allocation (HRTPA) problem, which involves determining when to perform tasks and who should execute them to maximize efficiency while ensuring workers' physical fatigue remains within safe limits. The inclusion of fatigue constraints, combined with production dynamics, significantly increases the complexity of the HRTPA problem. Traditional fatigue-recovery models in HRTPA often rely on static, predefined hyperparameters. However, in practice, human fatigue sensitivity varies daily due to factors such as changed work conditions and insufficient sleep. To better capture this uncertainty, we treat fatigue-related parameters as inaccurate and estimate them online based on observed fatigue progression during production. To address these challenges, we propose PF-CD3Q, a safe reinforcement learning (safe RL) approach that integrates the particle filter with constrained dueling double deep Q-learning for real-time fatigue-predictive HRTPA. Specifically, we first develop PF-based estimators to track human fatigue and update fatigue model parameters in real-time. These estimators are then integrated into CD3Q by making task-level fatigue predictions during decision-making and excluding tasks that exceed fatigue limits, thereby constraining the action space and formulating the problem as a constrained Markov decision process (CMDP).

Recent advances in 3D Gaussian Splatting (3DGS) have enabled impressive real-time photorealistic rendering. However, conventional training pipelines inherently assume full multi-view consistency among input images, which makes them sensitive to distractors that violate this assumption and cause visual artifacts. In this work, we revisit an underexplored aspect of 3DGS: its inherent ability to suppress inconsistent signals. Building on this insight, we propose PDF-GS (Progressive Distractor Filtering for Robust 3D Gaussian Splatting), a framework that amplifies this self-filtering property through a progressive multi-phase optimization. The progressive filtering phases gradually remove distractors by exploiting discrepancy cues, while the following reconstruction phase restores fine-grained, view-consistent details from the purified Gaussian representation. Through this iterative refinement, PDF-GS achieves robust, high-fidelity, and distractor-free reconstructions, consistently outperforming baselines across diverse datasets and challenging real-world conditions. Moreover, our approach is lightweight and easily adaptable to existing 3DGS frameworks, requiring no architectural changes or additional inference overhead, leading to a new state-of-the-art performance. The code is publicly available at https://github.com/kangrnin/PDF-GS.

Large language models (LLMs) face significant challenges in processing long contexts due to the linear growth of the key-value (KV) cache and quadratic complexity of self-attention. Existing approaches address these bottlenecks separately: Multi-head Latent Attention (MLA) reduces the KV cache by projecting tokens into a low-dimensional latent space, while sparse attention reduces computation. However, sparse methods cannot operate natively on MLA's compressed latent structure, missing opportunities for joint optimization. In this paper, we propose Latent-Condensed Attention (LCA), which directly condenses context within MLA's latent space, where the representation is disentangled into semantic latent vectors and positional keys. LCA separately aggregates semantic vectors via query-aware pooling and preserves positional keys via anchor selection. This approach jointly reduces both computational cost and KV cache without adding parameters. Beyond MLA, LCA's design is architecture-agnostic and readily extends to other attention mechanisms such as GQA. Theoretically, we prove a length-independent error bound. Experiments show LCA achieves up to 2.5$\times$ prefilling speedup and 90% KV cache reduction at 128K context while maintaining competitive performance.

Simulating Standardized Patients with cognitive impairment offers a scalable and ethical solution for clinical training. However, existing methods rely on discrete prompt engineering and fail to capture the heterogeneity of deficits across varying domains and severity levels. To address this limitation, we propose StsPatient for the fine-grained simulation of cognitively impaired patients. We innovatively capture domain-specific features by extracting steering vectors from contrastive pairs of instructions and responses. Furthermore, we introduce a Stochastic Token Modulation (STM) mechanism to regulate the intervention probability. STM enables precise control over impairment severity while mitigating the instability of conventional vector methods. Comprehensive experiments demonstrate that StsPatient significantly outperforms baselines in both clinical authenticity and severity controllability.

We introduce Context Kubernetes, an architecture for orchestrating enterprise knowledge in agentic AI systems, with a prototype implementation and eight experiments. The core observation is that delivering the right knowledge, to the right agent, with the right permissions, at the right freshness -- across an entire organization -- is structurally analogous to the container orchestration problem Kubernetes solved a decade ago. We formalize six core abstractions, a YAML-based declarative manifest for knowledge-architecture-as-code, a reconciliation loop, and a three-tier agent permission model where agent authority is always a strict subset of human authority. On synthetic seed data, we compare four governance baselines of increasing strength: ungoverned RAG, ACL-filtered retrieval, RBAC-aware routing, and the full architecture. Each layer contributes a different capability: ACL filtering eliminates cross-domain leaks, intent routing reduces noise by 19 percentage points, and only the three-tier model blocks all five tested attack scenarios -- the one attack RBAC misses is an agent sending confidential pricing via email, which RBAC cannot distinguish from ordinary email. TLA+ model-checking verifies safety properties across 4.6 million reachable states with zero violations. A survey of four major platforms (Microsoft, Salesforce, AWS, Google) documents that none architecturally isolates agent approval channels. We identify four properties that make context orchestration harder than container orchestration, and argue these make the solution more valuable.

Multimodal Large Language Models (MLLMs) have achieved remarkable progress but continue to struggle with geometric reasoning, primarily due to the perception bottleneck regarding fine-grained visual elements. While formal languages have aided plane geometry understanding, solid geometry which requires spatial understanding remains largely unexplored. In this paper, we address this challenge by designing a unified formal language that integrates plane and solid geometry, comprehensively covering geometric structures and semantic relations. We construct GDP-29K, a large-scale dataset comprising 20k plane and 9k solid geometry samples collected from diverse real-world sources, each paired with its ground-truth formal description. To ensure syntactic correctness and geometric consistency, we propose a training paradigm that combines Supervised Fine-Tuning with Reinforcement Learning via Verifiable Rewards. Experiments show that our approach achieves state-of-the-art parsing performance. Furthermore, we demonstrate that our parsed formal descriptions serve as a critical cognitive scaffold, significantly boosting MLLMs' capabilities for downstream geometry reasoning tasks. Our data and code are available at Geoparsing.

Fine-tuning pretrained image classifiers is standard practice, yet which individual samples are forgotten during this process, and whether forgetting patterns are stable or architecture dependent, remains unclear. Understanding these dynamics has direct implications for curriculum design, data pruning, and ensemble construction. We track per-sample correctness at every epoch during fine-tuning of ResNet-18 and DeiT-Small on a retinal OCT dataset (7 classes, 56:1 imbalance) and CUB-200-2011 (200 bird species), fitting Ebbinghaus-style exponential decay curves to each sample's retention trace. Five findings emerge. First, the two architectures forget fundamentally different samples: Jaccard overlap of the top 10 percent most-forgotten is 0.34 on OCTDL and 0.15 on CUB-200. Second, ViT forgetting is more structured (mean $R^2 = 0.74$) than CNN forgetting ($R^2 = 0.52$). Third, per-sample forgetting is stochastic across random seeds (Spearman $ρ\approx 0.01$), challenging the assumption that sample difficulty is an intrinsic property. Fourth, class-level forgetting is consistent and semantically interpretable: visually similar species are forgotten most, distinctive ones least. Fifth, a sample's loss after head warmup predicts its long-term decay constant ($ρ= 0.30$ to $0.50$, $p < 10^{-45}$). These findings suggest that architectural diversity in ensembles provides complementary retention coverage, and that curriculum or pruning methods based on per-sample difficulty may not generalize across runs. A spaced repetition sampler built on these decay constants does not outperform random sampling, indicating that static scheduling cannot exploit unstable per-sample signals.

Dual-encoder Vision-Language Models (VLMs) such as CLIP are often characterized as bag-of-words systems due to their poor performance on compositional benchmarks. We argue that this limitation may stem less from deficient representations than from the standard inference protocol based on global cosine similarity. First, through controlled diagnostic experiments, we show that explicitly enforcing fine-grained region-segment alignment at inference dramatically improves compositional performance without updating pretrained encoders. We then introduce a lightweight transformer that learns such alignments directly from frozen patch and token embeddings. Comparing against full fine-tuning and prior end-to-end compositional training methods, we find that although these approaches improve in-domain retrieval, their gains do not consistently transfer under distribution shift. In contrast, learning localized alignment over frozen representations matches full fine-tuning on in-domain retrieval while yielding substantial improvements on controlled out-of-domain compositional benchmarks. These results identify global embedding matching as a key bottleneck in dual-encoder VLMs and highlight the importance of alignment mechanisms for robust compositional generalization.

Developing non-collaborative dialogue agents traditionally requires the manual, unscalable codification of expert strategies. We propose \ours, a method that leverages large language models to autonomously induce both strategy actions and planning logic directly from raw transcripts. METRO formalizes expert knowledge into a Strategy Forest, a hierarchical structure that captures both short-term responses (nodes) and long-term strategic foresight (branches). Experimental results across two benchmarks show that METRO demonstrates promising performance, outperforming existing methods by an average of 9%-10%. Our further analysis not only reveals the success behind METRO (strategic behavioral diversity and foresight), but also demonstrates its robust cross-task transferability. This offers new insights into building non-collaborative agents in a cost-effective and scalable way. Our code is available at https://github.com/Humphrey-0125/METRO.

Accurate air traffic prediction in the terminal airspace (TA) is pivotal for proactive air traffic management (ATM). However, existing data-driven approaches predominantly rely on time series-based forecasting paradigms, which inherently overlook critical aircraft state information, such as real-time kinematics and proximity to airspace boundaries. To address this limitation, we propose \textit{AeroSense}, a direct state-to-flow modeling framework for air traffic prediction. Unlike classical time series-based methods that first aggregate aircraft trajectories into macroscopic flow sequences before modeling, AeroSense explicitly represents the real-time airspace situation as \textit{a dynamic set of aircraft states}, enabling the direct processing of a variable number of aircraft instead of time series as inputs. Specifically, we introduce a situation-aware state representation that enables AeroSense to sense the instantaneous terminal airspace situation directly from microscopic aircraft states. Furthermore, we design a model architecture that incorporates masked self-attention to capture inter-aircraft interactions, together with two decoupled prediction heads to model heterogeneous flow dynamics across two key functional areas of the TA. Extensive experiments on a large-scale real-world airport dataset demonstrate that AeroSense consistently achieves state-of-the-art performance, validating that direct modeling of microscopic aircraft states yields substantially higher predictive fidelity than time series-based baselines. Moreover, the proposed framework exhibits superior robustness during peak traffic periods, achieves Pareto-optimal performance under dayparting multi-object evaluation, and provides meaningful interpretability through attention-based visualizations.

The biological definition of Alzheimer's disease (AD) relies on multi-modal neuroimaging, yet the clinical utility of positron emission tomography (PET) is limited by cost and radiation exposure, hindering early screening at preclinical or prodromal stages. While generative models offer a promising alternative by synthesizing PET from magnetic resonance imaging (MRI), achieving subject-specific precision remains a primary challenge. Here, we introduce DIReCT$++$, a Domain-Informed ReCTified flow model for synthesizing multi-tracer PET from MRI combined with fundamental clinical information. Our approach integrates a 3D rectified flow architecture to capture complex cross-modal and cross-tracer relationships with a domain-adapted vision-language model (BiomedCLIP) that provides text-guided, personalized generation using clinical scores and imaging knowledge. Extensive evaluations on multi-center datasets demonstrate that DIReCT$++$ not only produces synthetic PET images ($^{18}$F-AV-45 and $^{18}$F-FDG) of superior fidelity and generalizability but also accurately recapitulates disease-specific patterns. Crucially, combining these synthesized PET images with MRI enables precise personalized stratification of mild cognitive impairment (MCI), advancing a scalable, data-efficient tool for the early diagnosis and prognostic prediction of AD. The source code will be released on https://github.com/ladderlab-xjtu/DIReCT-PLUS.

Customer service chatbots are increasingly expected to serve not merely as reactive support tools for users, but as strategic interfaces for harvesting high-value information and business intelligence. In response, we make three main contributions. 1) We introduce and define a novel task of Proactive Information Probing, which optimizes when to probe users for pre-specified target information while minimizing conversation turns and user friction. 2) We propose PROCHATIP, a proactive chatbot framework featuring a specialized conversation strategy module trained to master the delicate timing of probes. 3) Experiments demonstrate that PROCHATIP significantly outperforms baselines, exhibiting superior capability in both information probing and service quality. We believe that our work effectively redefines the commercial utility of chatbots, positioning them as scalable, cost-effective engines for proactive business intelligence. Our code is available at https://github.com/SCUNLP/PROCHATIP.