QUOKA: Query-Oriented KV Selection For Efficient LLM Prefill

We present QUOKA: Query-oriented KV selection for efficient attention, a training-free and hardware agnostic sparse attention algorithm for accelerating transformer inference under chunked prefill. While many queries focus on a smaller group of keys in the attention operator, we observe that queries with low cosine similarity with respect to the mean query interact more strongly with more keys and have the greatest contribution to final attention logits. By prioritizing these low cosine similarity queries, the behavior of full attention during the prefill stage can be closely approximated. QUOKA leverages this observation, accelerating attention by (1) first retaining a small set of representative queries and (2) then subselectin the keys most aligned with those queries. Through experiments on Needle-In-A-Haystack, LongBench, RULER, and Math500, we show that, while realizing a 3x reduction in time-to-first-token, 5x speedup in attention on Nvidia GPUs and up to nearly a 7x speedup on Intel Xeon CPUs, QUOKA achieves near-baseline accuracy, utilizing 88% fewer key-value pairs per attention evaluation.

Technosocial risks of ideal emotion recognition technologies: A defense of the (social) value of emotional expressions

The prospect of AI systems that I call ideal emotion recognition technologies (ERTs) is often defended on the assumption that social life would benefit from increased affective transparency. This paper challenges that assumption by examining the technosocial risks posed by ideal ERTs, understood as multimodal systems capable of reliably inferring inner affective states in real time. Drawing on philosophical accounts of emotional expression and social practice, as well as empirical work in affective science and social psychology, I argue that the appeal of such systems rests on a misunderstanding of the social functions of emotional expression. Emotional expressions function not only as read-outs of inner states, but also as tools for coordinating action, enabling moral repair, sustaining interpersonal trust, and supporting collective norms. These functions depend on a background of partial opacity and epistemic friction. When deployed in socially authoritative or evaluative contexts, ideal ERTs threaten this expressive space by collapsing epistemic friction, displacing relational meaning with technology-mediated affective profiles, and narrowing the space for aspirational and role-sensitive expressions. The result is a drift towards affective determinism and ambient forms of affective auditing, which undermine both social cohesion and individual agency. I argue that, although it is intuitive to think that increasing accuracy would legitimise such systems, in the case of ERTs accuracy does not straightforwardly justify their deployment, and may, in some contexts, provide a reason for regulatory restraint. I conclude by defending a function-first regulatory approach that treats expressive discretion and intentional emotional expression as constitutive of certain social goods, and that accordingly seeks to protect these goods from excessive affective legibility.

Learning To Sample From Diffusion Models Via Inverse Reinforcement Learning

Diffusion models generate samples through an iterative denoising process, guided by a neural network. While training the denoiser on real-world data is computationally demanding, the sampling procedure itself is more flexible. This adaptability serves as a key lever in practice, enabling improvements in both the quality of generated samples and the efficiency of the sampling process. In this work, we introduce an inverse reinforcement learning framework for learning sampling strategies without retraining the denoiser. We formulate the diffusion sampling procedure as a discrete-time finite-horizon Markov Decision Process, where actions correspond to optional modifications of the sampling dynamics. To optimize action scheduling, we avoid defining an explicit reward function. Instead, we directly match the target behavior expected from the sampler using policy gradient techniques. We provide experimental evidence that this approach can improve the quality of samples generated by pretrained diffusion models and automatically tune sampling hyperparameters.

SA-CAISR: Stage-Adaptive and Conflict-Aware Incremental Sequential Recommendation

Sequential recommendation (SR) aims to predict a user's next action by learning from their historical interaction sequences. In real-world applications, these models require periodic updates to adapt to new interactions and evolving user preferences. While incremental learning methods facilitate these updates, they face significant challenges. Replay-based approaches incur high memory and computational costs, and regularization-based methods often struggle to discard outdated or conflicting knowledge. To overcome these challenges, we propose SA-CAISR, a Stage-Adaptive and Conflict-Aware Incremental Sequential Recommendation framework. As a buffer-free framework, SA-CAISR operates using only the old model and new data, directly addressing the high costs of replay-based techniques. SA-CAISR introduces a novel Fisher-weighted knowledge-screening mechanism that dynamically identifies outdated knowledge by estimating parameter-level conflicts between the old model and new data, allowing our approach to selectively remove obsolete knowledge while preserving compatible historical patterns. This dynamic balance between stability and adaptability allows our method to achieve a new state-of-the-art performance in incremental SR. Specifically, SA-CAISR improves Recall@20 by 2.0%, MRR@20 by 1.2%, and NDCG@20 by 1.4% on average across datasets, while reducing memory usage by 97.5% and training time by 46.9% compared to the best baselines. This efficiency allows real-world systems to rapidly update user profiles with minimal computational overhead, ensuring more timely and accurate recommendations.

High-Speed Vision-Based Flight in Clutter with Safety-Shielded Reinforcement Learning

Quadrotor unmanned aerial vehicles (UAVs) are increasingly deployed in complex missions that demand reliable autonomous navigation and robust obstacle avoidance. However, traditional modular pipelines often incur cumulative latency, whereas purely reinforcement learning (RL) approaches typically provide limited formal safety guarantees. To bridge this gap, we propose an end-to-end RL framework augmented with model-based safety mechanisms. We incorporate physical priors in both training and deployment. During training, we design a physics-informed reward structure that provides global navigational guidance. During deployment, we integrate a real-time safety filter that projects the policy outputs onto a provably safe set to enforce strict collision-avoidance constraints. This hybrid architecture reconciles high-speed flight with robust safety assurances. Benchmark evaluations demonstrate that our method outperforms both traditional planners and recent end-to-end obstacle avoidance approaches based on differentiable physics. Extensive experiments demonstrate strong generalization, enabling reliable high-speed navigation in dense clutter and challenging outdoor forest environments at velocities up to 7.5m/s.

Deep Learning-Based Fixation Type Prediction for Quality Assurance in Digital Pathology

Accurate annotation of fixation type is a critical step in slide preparation for pathology laboratories. However, this manual process is prone to errors, impacting downstream analyses and diagnostic accuracy. Existing methods for verifying formalin-fixed, paraffin-embedded (FFPE), and frozen section (FS) fixation types typically require full-resolution whole-slide images (WSIs), limiting scalability for high-throughput quality control. We propose a deep-learning model to predict fixation types using low-resolution, pre-scan thumbnail images. The model was trained on WSIs from the TUM Institute of Pathology (n=1,200, Leica GT450DX) and evaluated on a class-balanced subset of The Cancer Genome Atlas dataset (TCGA, n=8,800, Leica AT2), as well as on class-balanced datasets from Augsburg (n=695 [392 FFPE, 303 FS], Philips UFS) and Regensburg (n=202, 3DHISTECH P1000). Our model achieves an AUROC of 0.88 on TCGA, outperforming comparable pre-scan methods by 4.8%. It also achieves AUROCs of 0.72 on Regensburg and Augsburg slides, underscoring challenges related to scanner-induced domain shifts. Furthermore, the model processes each slide in 21 ms, $400\times$ faster than existing high-magnification, full-resolution methods, enabling rapid, high-throughput processing. This approach provides an efficient solution for detecting labelling errors without relying on high-magnification scans, offering a valuable tool for quality control in high-throughput pathology workflows. Future work will improve and evaluate the model's generalisation to additional scanner types. Our findings suggest that this method can increase accuracy and efficiency in digital pathology workflows and may be extended to other low-resolution slide annotations.

Projected Gradient Ascent for Efficient Reward-Guided Updates with One-Step Generative Models

We propose a constrained latent optimization method for reward-guided generation that preserves white Gaussian noise characteristics with negligible overhead. Test-time latent optimization can unlock substantially better reward-guided generations from pretrained generative models, but it is prone to reward hacking that degrades quality and also too slow for practical use. In this work, we make test-time optimization both efficient and reliable by replacing soft regularization with hard white Gaussian noise constraints enforced via projected gradient ascent. Our method applies a closed-form projection after each update to keep the latent vector explicitly noise-like throughout optimization, preventing the drift that leads to unrealistic artifacts. This enforcement adds minimal cost: the projection matches the $O(N \log N)$ complexity of standard algorithms such as sorting or FFT and does not practically increase wall-clock time. In experiments, our approach reaches a comparable Aesthetic Score using only 30% of the wall-clock time required by the SOTA regularization-based method, while preventing reward hacking.

LEFT: Learnable Fusion of Tri-view Tokens for Unsupervised Time Series Anomaly Detection

As a fundamental data mining task, unsupervised time series anomaly detection (TSAD) aims to build a model for identifying abnormal timestamps without assuming the availability of annotations. A key challenge in unsupervised TSAD is that many anomalies are too subtle to exhibit detectable deviation in any single view (e.g., time domain), and instead manifest as inconsistencies across multiple views like time, frequency, and a mixture of resolutions. However, most cross-view methods rely on feature or score fusion and do not enforce analysis-synthesis consistency, meaning the frequency branch is not required to reconstruct the time signal through an inverse transform, and vice versa. In this paper, we present Learnable Fusion of Tri-view Tokens (LEFT), a unified unsupervised TSAD framework that models anomalies as inconsistencies across complementary representations. LEFT learns feature tokens from three views of the same input time series: frequency-domain tokens that embed periodicity information, time-domain tokens that capture local dynamics, and multi-scale tokens that learns abnormal patterns at varying time series granularities. By learning a set of adaptive Nyquist-constrained spectral filters, the original time series is rescaled into multiple resolutions and then encoded, allowing these multi-scale tokens to complement the extracted frequency- and time-domain information. When generating the fused representation, we introduce a novel objective that reconstructs fine-grained targets from coarser multi-scale structure, and put forward an innovative time-frequency cycle consistency constraint to explicitly regularize cross-view agreement. Experiments on real-world benchmarks show that LEFT yields the best detection accuracy against SOTA baselines, while achieving a 5x reduction on FLOPs and 8x speed-up for training.

Debate is efficient with your time

AI safety via debate uses two competing models to help a human judge verify complex computational tasks. Previous work has established what problems debate can solve in principle, but has not analysed the practical cost of human oversight: how many queries must the judge make to the debate transcript? We introduce Debate Query Complexity}(DQC), the minimum number of bits a verifier must inspect to correctly decide a debate. Surprisingly, we find that PSPACE/poly (the class of problems which debate can efficiently decide) is precisely the class of functions decidable with O(log n) queries. This characterisation shows that debate is remarkably query-efficient: even for highly complex problems, logarithmic oversight suffices. We also establish that functions depending on all their input bits require Omega(log n) queries, and that any function computable by a circuit of size s satisfies DQC(f) <= log(s) + 3. Interestingly, this last result implies that proving DQC lower bounds of log(n) + 6 for languages in P would yield new circuit lower bounds, connecting debate query complexity to central questions in circuit complexity.

CauScale: Neural Causal Discovery at Scale

Causal discovery is essential for advancing data-driven fields such as scientific AI and data analysis, yet existing approaches face significant time- and space-efficiency bottlenecks when scaling to large graphs. To address this challenge, we present CauScale, a neural architecture designed for efficient causal discovery that scales inference to graphs with up to 1000 nodes. CauScale improves time efficiency via a reduction unit that compresses data embeddings and improves space efficiency by adopting tied attention weights to avoid maintaining axis-specific attention maps. To keep high causal discovery accuracy, CauScale adopts a two-stream design: a data stream extracts relational evidence from high-dimensional observations, while a graph stream integrates statistical graph priors and preserves key structural signals. CauScale successfully scales to 500-node graphs during training, where prior work fails due to space limitations. Across testing data with varying graph scales and causal mechanisms, CauScale achieves 99.6% mAP on in-distribution data and 84.4% on out-of-distribution data, while delivering 4-13,000 times inference speedups over prior methods. Our project page is at https://github.com/OpenCausaLab/CauScale.