arXiv:2607.16381v1 Announce Type: new
Abstract: We develop a deterministic, scale-resolved formulation of energy transfer in the three-dimensional incompressible Navier-Stokes equations based on an explicit triadic decomposition of the nonlinear term in Fourier space. Using a systematic dyadic localization of the velocity field, we derive an exact representation of the nonlinear energy flux across scales and organize it in terms of interactions between well-defined scale components. Under suitable smoothness assumptions, we obtain an absolutely convergent triadic expansion and quantitative bounds that distinguish local and nonlocal contributions in scale space. This framework provides a transparent and fully explicit description of how energy transfer is mediated by triadic interactions and how scale locality emerges as a structural property of the nonlinearity. Building on this formulation, we revisit the classical inertial-range picture of turbulence from a deterministic perspective. We show that, under a scale-invariant flux assumption, the Kolmogorov $-5/3$ scaling is formally consistent with the triadic energy-transfer mechanism at a structural level. The result does not rely on statistical assumptions, but instead follows from the structural properties of the Navier-Stokes equations combined with a scale-resolved representation of the energy flux. The present work thus provides a coherent synthesis of triadic interaction analysis, dyadic scale decomposition, and classical turbulence phenomenology, offering a deterministic framework that clarifies how Kolmogorov-type scaling constraints arise in the scale-resolved structure of the underlying equations.
Science Journals
arXiv:2607.17684v1 Announce Type: cross
Abstract: We study universal monotonicity and Frank--Wolfe stability properties for atomic splittable congestion games. Specifically, we characterize the largest resource cost class for which the associated variational inequality operator is monotone for every game. This characterization is given by a curvature inequality involving the first two derivatives of the allowable cost functions and the number of players. Our framework yields exact characterizations for universal monotonicity, strict monotonicity, and strong monotonicity; the strict and strong variants require corresponding stricter curvature conditions. We then draw a perhaps surprising connection to learning dynamics in atomic splittable congestion games. We show that the very same curvature condition also characterizes universal \emph{local and global stability} of the Euclidean-regularized Frank--Wolfe dynamics on arbitrary convex strategy spaces, provided the cost class is closed under positive affine transformations. Finally, we study games on simplices and show that an interior equilibrium of the regularized Frank--Wolfe dynamic is locally exponentially stable, even without the curvature condition.
arXiv:2607.16546v1 Announce Type: new
Abstract: Affective behavior recognition in the wild requires joint prediction of continuous valence-arousal, categorical facial expression, and multi-label action units from unconstrained face images. We present our system for the Multi-Task Learning (MTL) track of the 11th Affective Behavior Analysis in-the-wild (ABAW) competition on s-Aff-Wild2, the static selected-frame version of Aff-Wild2. The method focuses on post-encoder adaptation: frozen AffectNet-supervised backbones provide multi-resolution features, while task-specific temporal heads and cross-task fusion modules select the useful signals for each target. For action-unit recognition, we adapt MAE-Face with Low-Rank Adaptation (LoRA) and use DISFA through per-unit expert routing rather than direct sequential transfer. Ablations over backbone, temporal, fusion, and AU-adaptation choices define the final configuration. The final system obtains P = 1.7302 on the official validation split, showing that post-encoder adaptation and task-wise modeling choices provide a strong MTL pipeline without training a new large-scale face foundation model.
arXiv:2607.16267v1 Announce Type: new
Abstract: Continual learning in visual navigation remains challenging due to catastrophic forgetting and the difficulties associated with adapting to diverse and evolving environments. To address these issues, we propose Hyperbolic Dynamic Cluster Memory (HyperDCM), a structure-aware memory mechanism that enhances diffusion policy-based navigation through scene graph modeling and principled memory replay. HyperDCM extracts semantic scene triples from RGB observations using large vision-language models, encodes them into scene graph embeddings via a Relational Graph Convolutional Network (R-GCN), and projects the embeddings into hyperbolic space to enhance structural separability and retention in continual navigation. A dynamic clustering and structure-sensitive update strategy selects representative samples for memory replay, thereby preserving knowledge diversity and mitigating catastrophic forgetting. Experiments on multi-scene indoor and outdoor datasets demonstrate that HyperDCM achieves superior retention of past navigation capabilities and improved generalization compared to representative continual learning baselines adapted to diffusion policy navigation.
arXiv:2607.17058v1 Announce Type: new
Abstract: Precise metric depth estimation is fundamental for autonomous robot navigation, yet monocular systems inherently suffer from scale ambiguity and scale drift. While recent recurrent flow-based SLAM systems have demonstrated state-of-the-art robustness, they remain scale-ambiguous. In this paper, we propose Metric-DROID, an end-to-end recurrent architecture that anchors visual SLAM to physical reality by integrating proprioceptive odometry. Our framework introduces the following innovations: (1) A LSTM Update Operator that encodes high-frequency odometry sequences into spatial feature maps, providing a persistent metric bias for iterative refinement. (2) An Uncertainty-Aware Metric Backend ($BA_{odom}$) that treats odometry as a geometric anchor with learned heteroscedastic covariance. By regressing a time-varying metric uncertainty $\Sigma_{o}$, our system intelligently balances visual re-projection and metric translation residuals, effectively mitigating the impact of wheel-slip and sensor noise. (3) We further propose a selective residual fine-tuning strategy to preserve pre-trained geometric priors while enabling zero-shot metric alignment.
arXiv:2607.16673v1 Announce Type: new
Abstract: Linear-attention models replace the growing KV cache with recurrent states, but autoregressive decoding still reads, updates, and writes these states one token at a time. Speculative decoding can reduce this cost by verifying several draft tokens in one target pass, yet existing speculative systems are designed for Transformer KV caches. For stateful linear-attention targets, verification must follow recurrent dependencies across chains and branches, acceptance must update only the accepted state trajectory, and the drafter must avoid submitting candidates that waste stateful verification work. This paper presents SpecLA, a speculative decoding runtime for stateful linear-attention models. SpecLA verifies chains and trees with topology-aware kernels, stores compact factors produced during verification to recover accepted states, and uses confidence pruning plus a target-aligned EAGLE-style drafter to feed useful candidates to the verifier. On an NVIDIA H100 with a public GDN-1.3B target, SpecLA achieves up to 1.70x end-to-end speedup over autoregressive decoding.
arXiv:2607.17060v1 Announce Type: new
Abstract: A Bayes-filtered transformer (BFT) is a transformer trained on sequences that are generated in two steps: first a latent task is drawn from a prior, then observations are drawn conditional on that task. Trained under autoregressive log loss, the BFT's next-token prediction, in the idealized limit, is the Bayesian posterior predictive distribution (PPD) induced by that prior and that conditional law. In practice the trained BFT is only an approximation of this ideal PPD, raising an interpretive question: what prior and posterior over the latent task has the trained BFT actually internalized? Existing work answers this question by comparing the trained BFT's predictions against the predictions of various "reference" posteriors, each standing in for a different candidate algorithm or computation the BFT might be implementing. This prediction-space comparison is fragile: different posteriors can share the same posterior-mean predictions. We use predictive Monte Carlo (PMC) as a general interpretability tool for any BFT: using only next-token generation, PMC returns an approximation to the implicit prior and posterior over the latent task, answering the interpretive question directly in latent space. We apply PMC to three stylized task families spanning 0-Markov and 1-Markov exchangeability. The phenomena previously reported in these settings remain visible in latent space. Code is available at https://github.com/afiq-aswadi/bft-pmc
arXiv:2607.16674v1 Announce Type: new
Abstract: Accurate routability estimation during physical design is important for reducing costly post-routing iterations. Prior learning-based methods treat this task as deterministic prediction, mapping placement-stage features to a single congestion or DRC outcome. We instead formulate routability estimation as a conditional generation problem, where both routing congestion and DRC violations are modeled as spatially structured routability fields. Our framework, Conditional Latent Diffusion for Routeability estimation (CLDRoute), uses physics-aware conditioning and task-specific latent modeling to handle the different characteristics of congestion and DRC maps. This allows our method to supports sample-based inference, producing both a mean prediction and a spatial uncertainty estimate for the same input design. On CircuitNet 2.0 (N28), our method achieves, for DRC violation generation, an SSIM of 0.9678, an MAE of 0.0028, and a TopK@1% of 0.3494; for congestion generation, it achieves an SSIM of 0.9031, an MAE of 0.0286, and an NZ-Pearson of 0.3692. Overall, our framework provides a more practical view of routability at placement by generating both the expected outcome and its uncertainty.
arXiv:2607.16384v1 Announce Type: new
Abstract: For stochastic gradient descent (SGD) with a constant stepsize $\alpha$, the invariant law of the iterates, centered at a minimizer, describes the behavior of the algorithm over long time horizons. In the strongly convex case, this invariant law has the familiar $\sqrt{\alpha}$ scaling and a Gaussian limit as $\alpha\downarrow 0$. We show that this behavior changes fundamentally for convex objectives $H$ with flat minima and (sub)quadratic tails.
More specifically, we study SGD with Markovian noise generated by a contractive driving chain. For every sufficiently small constant stepsize $\alpha$, we prove existence, uniqueness, and geometric convergence to an augmented invariant law in a Wasserstein distance induced by an $\alpha$-dependent metric. When the minimizer $x_\star$ has local flatness exponent $m\ge2$, meaning that $\nabla^2 H(x)\asymp \lVert x-x_\star\rVert^{m-2} I_d$ as $x\to x_\star$, we obtain a contraction bound with factor $1-c\alpha^{m-1}$, where $c>0$ is a constant. This recovers the factor $1-c\alpha$ in the quadratic case $m=2$. We then analyze the small-stepsize scaling limit. We show that the invariant law concentrates on the scale $\alpha^{1/m}$ and that the rescaled iterates converge weakly to the stationary distribution of the stochastic differential equation $$
dY_t=-h_0(Y_t)\,dt+\Sigma^{1/2}\,dB_t , $$ where $h_0$ is the limiting drift at the minimizer and $\Sigma$ denotes the asymptotic covariance. This recovers the Gaussian limit when $m=2$ and gives generally non-Gaussian stationary limits in the flat case $m>2$. Finally, we give corresponding results for coordinate-separable objectives with unequal flatness exponents.
arXiv:2607.17340v1 Announce Type: new
Abstract: Domain-incremental object detection (DIOD) requires models to continually adapt to new domains while preserving prior knowledge. Recently, parameter-efficient fine-tuning offers a promising avenue, wherein a pre-trained model is frozen and a small number of learnable parameters are injected for downstream tasks. However, these methods risk overwriting critical past knowledge, triggering inter-domain interference and performance degradation. To address this challenge, we propose Orthogonal Knowledge Refreshing (OKR), a simple yet effective framework for DIOD. OKR incrementally constructs independent domain-specific subspaces via dedicated low-rank branches for each domain, which are seamlessly fused for a holistic decision, enabling conflict-free capacity expansion without domain selection during inference. To minimize knowledge interference during fusion, we present a gradient-based orthogonal refreshing strategy that projects gradient updates of new domains onto the orthogonal complement of the fused historical subspace, supporting continual adaptation without forgetting. Moreover, to mitigate semantic fragmentation across domains, we enforce topology-aware consistency, aligning the semantic structures of old and new domains. Extensive experiments validate the superiority of OKR, outperforming the best exemplar-free method by significant margins of +5.6% and +6.5% mAP on the Pascal VOC and BDD100K series, respectively.
arXiv:2412.11475v3 Announce Type: replace
Abstract: We present OmniVLM, a sub-billion-parameter vision-language model for efficient on-device inference. OmniVLM introduces a token compression mechanism that reduces visual token sequence length from 729 to 81 tokens, significantly reducing computational overhead while preserving visual-semantic fidelity. Through a multi-stage training pipeline of pretraining, supervised fine-tuning, and minimal-edit Direct Preference Optimization (DPO), OmniVLM matches the performance of larger models. On multiple benchmarks including ScienceQA, POPE, and MMMU, OmniVLM outperforms existing baselines like nanoLLAVA within a 968M-parameter footprint. Empirical results on the same laptop demonstrate 9.1x faster time-to-first-token (0.75s vs 6.82s) and 1.5x higher decoding speed (29.41 vs 19.20 tokens/s) compared to nanoLLAVA, enabling efficient deployment on edge devices. The model weights can be accessed on huggingface: https://huggingface.co/NexaAIDev/OmniVLM-968M, and the inference examples can be find in Appendix B.
arXiv:2607.17067v1 Announce Type: new
Abstract: Generative AI (GenAI) is reshaping software engineering, raising concerns about how the development pathway through which juniors become seniors is being eroded. While macro statistics show a decline in junior hiring and controlled studies demonstrate the effects of AI on individual task performance, the mechanisms through which GenAI reshapes early-career development in real organizational and educational contexts have not been thoroughly examined. Through 14 semi-structured interviews with juniors at the threshold of entering software engineering and senior software engineers in South Korea, analyzed using Reflexive Thematic Analysis, we reveal a foundational pattern of Absorption -- GenAI redirects entry-level work into senior-AI workflows -- and three consequences: (1) juniors losing the productive struggle through which expertise once developed; (2) the structural reproduction of this loss through collective normalization of GenAI use in university classrooms; and (3) the perceptual asymmetry between seniors and juniors that prevents either side from correcting these dynamics on their own. By extending learning theory and situated cognition to organizational and institutional scales, we argue that GenAI appears to be absorbing not just specific categories of tasks but also parts of the pathway through which the next generation of seniors is formed. Preserving this pathway will require deliberate institutional design across classrooms, workplaces, and the evaluation criteria for juniors.
arXiv:2501.15303v4 Announce Type: replace
Abstract: We study the guarded negation fragment of transitive closure logic (GNTC). We show that the satisfiability problem for GNTC is 2ExpTime-complete, by establishing the following reductions: (i) a polynomial-time reduction from the satisfiability problem for GNTC to the satisfiability problem for the unary negation fragment UNTC of GNTC, and (ii) a direct exponential-time reduction from the satisfiability problem for UNTC to the non-emptiness problem for 2-way alternating parity tree automata. Furthermore, we show that the model checking problem for GNTC is $\mathsf{P}^{\mathsf{NP}[\mathcal{O}(\log^2 n)]}$-complete in combined complexity. Our result implies $\mathsf{P}^{\mathsf{NP}[\mathcal{O}(\log^2 n)]}$-completeness for both UNTC and $\mathrm{UNFO}^{\mathrm{reg}}$, which were left open in previous works.
arXiv:2607.16549v1 Announce Type: new
Abstract: Language models (LMs) are trained to excel at predicting the next word in the sequence given prior context, and humans also share this predictability in reading comprehension. Neuroscience research reveals that next-word predictability influences brain response, as recorded at millisecond resolution using electroencephalography (EEG). While our evidence indicates that advanced LMs achieve accuracies closely aligned with human performance at the next-word prediction task, this raises the question: Does higher prediction accuracy necessarily mean that these models adequately capture the cognitive signals associated with human reading comprehension? Here, we generate regressors for both humans and LMs based on two information measures, including top-1 prediction and surprisal, to predict event-related potential (ERP) elicited from EEG recordings which reflect different stages of cognitive processing during reading. We argue that modelling ERP patterns offers fine-grained analysis of the cognitive plausibility of various LMs during reading. Our results indicate that only surprisal potentially correlates with language-processing ERPs, especially for open-class words with high semantic content. Moreover, our findings challenge the assumption that scaling LMs with increased parameters and computational budgets will consistently lead to improved convergence with human-like linguistic processing.
arXiv:2607.16822v1 Announce Type: new
Abstract: This work proposes a compact and hardware-efficient frequency- and radiation-pattern-reconfigurable antenna (FPRA) for passive multi-target direction-of-arrival (DOA) estimation using a single receive RF chain. The antenna consists of a sectorized circular patch loaded with 16 PIN diodes. By switching the diode states, the current-concentration boundary on the patch is shifted, enabling reconfiguration of both the operating frequency and radiation pattern. With a single receive RF chain, the proposed antenna achieves beam scanning from -40 degrees to 40 degrees and provides multiple operating frequencies across the S- and C-bands. Based on these reconfigurable observation states, radiation-pattern switching is used to emulate the spatial sampling of a conventional antenna array, while multi-frequency observations introduce phase diversity to reduce the correlation among echoes from multiple passive targets illuminated by the same transmitter. Experimental results demonstrate that the combined virtual spatial sampling and frequency diversity enable passive multi-target DOA estimation without a conventional antenna array or multiple receive RF chains. The proposed FPRA offers a compact and hardware-efficient sensing solution for future integrated sensing and communication (ISAC) systems.
arXiv:2607.17551v1 Announce Type: new
Abstract: Lung ultrasound (LUS) is a bedside tool for assessing pulmonary edema in patients at risk due to heart failure or impaired kidney function. However, automated LUS analysis remains challenging because of speckle noise, imaging artifacts, and operator-dependent acquisition variability. In this work, we present a deep learning framework for multi-class LUS video classification that explores two components: hierarchy-aware training, and anatomy-guided learning. Starting from a strong baseline, we introduce hierarchical training strategies and then introduce pleural line mask supervision to guide model attention toward anatomically relevant regions. We study four clinically relevant classes--healthy, B-lines, consolidations, and mixed B-lines with consolidations--using an open-access dataset of 1,886 videos from 219 patients, evaluated with patient-level five-fold cross-validation. Results show that hierarchy-aware training improves pathological separation relative to flat classification, while mask-guided attention supervision achieves the highest mean macro-F1 of 65.7\% and produces more localized attention patterns. Transfer experiments on the external COVID-BLUeS dataset further show competitive and parameter-efficient adaptation while preserving pleural-focused attention behavior. These findings suggest that combining clinically structured objectives with anatomy-guided supervision is a practical approach to robust, interpretable LUS video analysis. Code and model implementations are available at https://github.com/Alya-Almsouti/LUS-video-classification.
arXiv:2501.15813v2 Announce Type: replace
Abstract: A systematic method is presented for transforming a pair of initial polarization states (SoPs) into a corresponding target pair using a fixed configuration of two quarter-waveplates and one half-waveplate. The transformation is modeled as a rigid-body rotation in the three-dimensional Stokes space, allowing analytical derivation of waveplate angles that realize the desired mapping. While most existing analytical approaches assume linear or orthogonal states, the proposed method supports arbitrary target SoPs, including elliptical and non-orthogonal configurations in the Stokes space, provided they share a common rigid-body rotation with the inputs. Although the derivation is formulated for a single pair of SoPs, the method naturally extends to multiple SoPs under the same rotation condition. This enables deterministic and scalable polarization control for applications such as quantum key distribution, polarization-based signal processing, and integrated photonic systems.
arXiv:2607.17140v1 Announce Type: new
Abstract: Unified multimodal models (UMMs) aim to integrate visual understanding and generation within a single architecture, but architectural unification alone does not ensure semantic consistency. A model may describe the intended target correctly while generating an inconsistent edit. This exposes an understanding-generation alignment gap: linguistic and visual outputs live in different spaces, yet should be governed by the same target semantics. We study this gap in image editing, where an instruction defines a target state that can be both described and visually realized. Given a source image and an edit instruction, we compare a UMM's target caption with its edited image to test whether the two outputs converge on the same result. Our analysis shows that existing UMMs remain weakly aligned, especially for fine-grained entities, attributes, spatial relations, and local details, indicating that semantic unification is not achieved by architecture alone. To bridge this gap, we propose STBridge, a shared-target alignment framework that connects understanding and generation through a common target state. Here the target caption expresses the desired visual result, while the edited image realizes it visually, replacing separate task-specific paths with a shared information flow from target expression to target realization. STBridge follows an align-then-optimize strategy: supervised fine-tuning first establishes the shared-target channel, and sequential reinforcement learning further refines target-centered coordination. Across visual understanding, image generation, and image editing benchmarks, STBridge consistently improves over the initialization model. Alignment analysis confirms that STBridge narrows the gap between what the model describes and what it generates, demonstrating shared-target alignment as an effective post-training strategy for bridging understanding and generation in UMMs.
arXiv:2607.17553v1 Announce Type: new
Abstract: This paper proposes a novel on-off division multiple access (ODMA) transmission scheme that enables efficient joint multi-user channel estimation and iterative decoding by inserting a small number of frozen symbols into the codewords. Functionally analogous to pilots, these symbols are sparsely distributed within the codeword. Unlike conventional pilot-based methods, our approach requires only a minimal number of frozen symbols (e.g., $5\sim20$ symbols per user in a 300-user system), which serve dual purposes as both estimation references and decoding aids. By employing low-complexity single-user channel estimation and decoding, combined with simple iterative interference cancellation, the scheme achieves performance equivalent to that with perfectly known user channels, even when accounting for the additional energy and bandwidth costs of the frozen symbols. Furthermore, for the large-scale ODMA sparse multiple access system, this paper proposes a fixed-point analysis method, which can accurately estimate the iterative convergence performance over multi-user block fading channels by only leveraging the decoding functions under the single-user AWGN channel. This method is applicable to performance analysis for arbitrary code lengths, code rates, and decoding algorithms. It eliminates the need for extensive Monte Carlo simulation time, and provides an efficient tool for the design of multi-user codes.
arXiv:2607.17555v1 Announce Type: new
Abstract: Predicting extreme price movements in high-frequency financial markets is a challenging task due to non-stationarity, heavy-tailed return distributions, and severe class imbalance. In particular, rare but impactful events are often difficult to detect using conventional modeling approaches, which typically treat extreme movements as isolated observations. This study proposes a volatility-aware approach for extreme event detection using high-frequency Bitcoin limit order book (LOB) data. Motivated by empirical evidence of volatility clustering, the target formulation is extended to incorporate both large future returns and high-volatility regimes. This redefinition increases the proportion of informative samples and aligns the learning objective with the underlying market dynamics. Using a tree-based model (XGBoost) with time-series cross-validation and imbalance-aware evaluation, the proposed method achieves a Precision-Recall AUC of approximately 0.40, significantly outperforming the baseline formulation with a PR-AUC of around 0.06. This represents more than a sixfold improvement in detecting rare events. The results highlight that target design plays a critical role in financial machine learning, often exceeding the impact of model complexity. By incorporating volatility structure into the labeling process, the proposed approach provides a more effective and realistic framework for extreme event detection in high-frequency cryptocurrency markets.
arXiv:2607.16360v1 Announce Type: new
Abstract: Over-the-air computation (AirComp) exploits multiple-access superposition to compute functions of distributed data without separately decoding all terminal messages. We study a reusable two-input AirComp gate for the exp-minus-log (EML) operation $\eml(u,v)=\exp(u)-\log(v)$, $v>0$. Thus, all internal nodes of a prescribed real-admissible EML tree reuse one gate type, avoiding node-specific nonlinear gate designs. Given an explicit EML tree whose intermediate logarithm arguments remain positive on a given compact domain, we derive additive white Gaussian noise (AWGN) and coherent flat fading implementations under peak-power constraints. We then characterize the number of gate evaluations, the dependency depth, evaluation latency, node-wise feasibility, deterministic error propagation, positivity preservation, and a high-probability AWGN error bound for the complete tree. A four-terminal two-hop example gives explicit positivity and end-to-end error conditions, and a digital interface propagates quantization and gate errors across the tree.
Dependency-Guided Code Generation: Structured Matrix Decomposition and Consistency-Guided Refinement
arXiv:2607.16692v1 Announce Type: new
Abstract: The increasing complexity of modern software systems has made automated code generation a fundamental task in software engineering. However, existing approaches often fail to adequately capture the intricate, multi-level dependencies among code entities, leading to generated code that is logically incomplete or difficult to integrate into real-world systems. To address this limitation, we propose a dependency-aware code generation framework that explicitly models interactions among code entities through a graph-based representation. We decompose dependencies into two complementary components: a quantized matrix that captures strong, explicit relations, and a sparse low-rank factorization that models weaker, implicit interactions. The decomposition is efficiently learned via an alternating optimization procedure. During code generation, the learned dependency structure is incorporated as a constraint, ensuring both semantic coherence and structural consistency of the generated code. Furthermore, we introduce a sparse triplet representation for strong dependencies, significantly improving storage efficiency and computational scalability. Extensive experiments demonstrate that our approach consistently produces code with superior semantic alignment and structural fidelity compared to existing methods.
arXiv:2607.17558v1 Announce Type: new
Abstract: On-policy self-distillation (OPSD) offers a promising approach for training large language models without relying on a separate teacher model. However, its effectiveness on complex agentic tasks remains largely unexplored. In this work, we instantiate Feedback-Augmented Self-Distillation (FA-SD), a self-distillation algorithm for agentic search that leverages successful demonstrations as privileged information. We identify that models can rely on recurring reasoning-and-search output templates, producing trajectories that appear diverse but are largely agnostic to the input question, making the KL-based self-distillation signal uninformative. We term this phenomenon decoding collapse, a failure mode that can be missed by existing evaluation metrics. To understand its underlying cause, we show that although the self-teacher achieves stronger performance, learning remains inherently unstable due to inconsistent supervision signals. We further decompose this inconsistency into model inconsistency and prompt inconsistency, and show that the latter can significantly degrade the quality of the supervision signal, limiting the effectiveness of self-teacher learning. To mitigate this inconsistency, we introduce an exponential moving average (EMA) teacher to stabilize the self-teacher and provide more consistent supervision signals. Although the EMA teacher requires a warm-up phase during which performance may temporarily regress, it ultimately improves model performance by providing more stable supervision.
arXiv:2607.17560v1 Announce Type: new
Abstract: Reinforcement learning (RL) provides a framework for sequential decision making under explicit objectives. In its classical form, RL studies how an agent should act to maximise long-term reward in a dynamic environment. In richer settings, the problem extends beyond a single agent and fixed environment: intelligent behavior may require strategic interaction, adaptation to uncertainty, and reasoning over high-dimensional worlds. This thesis studies RL from two perspectives: algorithms in games and RL in the era of foundation models.
The first part focuses on multi-agent RL in games. It examines how incentives, policies, and equilibrium concepts interact in competitive and general-sum environments, spanning two-player zero-sum games, large-scale video games, and multi-player settings with general structure. These works investigate learning in multi-agent systems and the behavior of RL methods in interactive environments. The second part studies RL with generative and foundation models, motivated by the idea that prior knowledge can enrich sequential decision making. Pretrained generative models and learned world models serve as representation tools and structured priors for planning, control, and policy optimization. The thesis develops diffusion-based world models, investigates RL for efficient video generation, explores generative models as policy classes, and studies interactive video world models in which actions shape future observations. It also addresses long-horizon modeling through architectures with memory. Together, these contributions present a unified view of RL as objective-driven adaptation in complex sequential domains. From strategic games to generative world models, the thesis highlights how RL connects decision making, environment modeling, and emerging foundation-model capabilities, offering a broader perspective on the principles underlying intelligent behavior.
arXiv:2607.17564v1 Announce Type: new
Abstract: AI companions are judged not only by single-turn fluency but by whether they sustain emotional continuity: remembering who the companion is, what the user prefers, and how the relationship has felt. We present ZifaMem, a structured memory system that organizes dialogue into session summaries, episodic memories, and a consolidated user model. Against a deployment-honest comparator that supplies the full raw dialogue history, and under a fixed LLM-as-a-judge protocol with route audits, structured memory raises pooled four-backbone emotional-intelligence scores by 11.4% (95% CI 6.3% to 17.1%), and persona grounding improves on all four backbones (Claude +42% relative). Multi-turn affect context wins a +39% net preference over a single-turn snapshot (exploratory), whereas an additional emotion state machine yields no measurable gain on any of five endpoints. Under an identical preregistered protocol, three memory systems (ZifaMem, Mem0, and filtered verbatim retrieval) each improve significantly over raw-history deployment, and ZifaMem and Mem0 are statistically equivalent within +/-5 points on the preregistered primary preference endpoint. The ZifaMem SDK, CLI, and portable Agent Skills are open-sourced at https://github.com/zifacorp/zifamem.