A common approach is to cast this problem in terms of a dynamical system with point attractor dynamics, where a small number of parameters drive the vocal tract to a stable equilibrium position (Browman and Goldstein, 1986; Fowler, 1980; Gafos, 2006; Saltzman and Munhall, 1989; Tilsen, 2016). A standard model in this framework is the linear harmonic oscillator, m x + b x + kx = 0 (1) where m is mass (typically m = 1), k is a stiffness coefficient, and b is a damping coefficient, usually set to critically damped b = 2 mk. Gestural activation can be governed by step activation, with gestural parameters changing instantaneously at the point of activation and remaining constant over the activation interval. In this study we focus on whether the parameters of a linear harmonic oscillator can be estimated from ultrasound tongue imaging data, which we compare with the more common method of fitting to electromagnetic articulography (EMA) data. A major barrier to this goal is that the linear harmonic oscillator is known to be a poor fit to empirical articulatory trajectories, as it predicts overly short time-to-peak velocity, meaning that it is inappropriate for evaluating how the model can fit different data modalities. There are three common solutions to this issue. The first allows gestural activation to vary over time (Byrd and Saltzman, 1998), which adds extrinsic complexity to the model. The second is a nonlinear model, such as adding a cubic term to the linear model (Kirkham, 2025b; 2 Sorensen and Gafos, 2016), or novel nonlinear models (Stern and Shaw, 2025). The third is to abandon oscillatory models and develop new time-dependent (i.e.

Nanopore sequencing enables real-time long-read DNA sequencing with reads exceeding 10 kilobases, but inherent error rates of 12-15 percent present significant computational challenges for read alignment. The critical seed chaining step must connect exact k-mer matches between reads and reference genomes while filtering spurious matches, yet state-of-the-art methods rely on fixed gap penalty functions unable to adapt to varying genomic contexts including tandem repeats and structural variants. This paper presents RawHash3, a hybrid framework combining graph neural networks with classical dynamic programming for adaptive seed chaining that maintains real-time performance while providing statistical guarantees. We formalize seed chaining as graph learning where seeds constitute nodes with 12-dimensional feature vectors and edges encode 8-dimensional spatial relationships including gap consistency. Our architecture employs three-layer EdgeConv GNN with confidence-based method selection that dynamically switches between learned guidance and algorithmic fallback. Comprehensive evaluation on 1,000 synthetic nanopore reads with 5,200 test seeds demonstrates RawHash3 achieves 99.94 percent precision and 40.07 percent recall, representing statistically significant 25.0 percent relative improvement over baseline with p less than 0.001. The system maintains median inference latency of 1.59ms meeting real-time constraints, while demonstrating superior robustness with 100 percent success rate under 20 percent label corruption versus baseline degradation to 30.3 percent. Cross-validation confirms stability establishing graph neural networks as viable approach for production genomics pipelines.

Parkinson's disease assessment has garnered growing interest in recent years, particularly with the advent of sensor data and machine learning techniques. Among these, multimodal approaches have demonstrated strong performance by effectively integrating complementary information from various data sources. However, two major limitations hinder their practical application: (1) the need to synchronize all modalities during training, and (2) the dependence on all modalities during inference. To address these issues, we propose the first Parkinson's assessment system that formulates multimodal learning as a multi-objective optimization (MOO) problem. This not only allows for more flexible modality requirements during both training and inference, but also handles modality collapse issue during multimodal information fusion. In addition, to mitigate the imbalance within individual modalities, we introduce a margin-based class rebalancing strategy to enhance category learning. We conduct extensive experiments on three public datasets under both synchronous and asynchronous settings. The results show that our framework-Towards Relaxed InPuts (TRIP)-achieves state-of-the-art performance, outperforming the best baselines by 16.48, 6.89, and 11.55 percentage points in the asynchronous setting, and by 4.86 and 2.30 percentage points in the synchronous setting, highlighting its effectiveness and adaptability.

Evaluating the naturalness of dialogue in language models (LMs) is not trivial: notions of 'naturalness' vary, and scalable quantitative metrics remain limited. This study leverages the linguistic notion of 'at-issueness' to assess dialogue naturalness and introduces a new method: Divide, Generate, Recombine, and Compare (DGRC). DGRC (i) divides a dialogue as a prompt, (ii) generates continuations for subparts using LMs, (iii) recombines the dialogue and continuations, and (iv) compares the likelihoods of the recombined sequences. This approach mitigates bias in linguistic analyses of LMs and enables systematic testing of discourse-sensitive behavior. Applying DGRC, we find that LMs prefer to continue dialogue on at-issue content, with this effect enhanced in instruct-tuned models. They also reduce their at-issue preference when relevant cues (e.g., "Hey, wait a minute") are present. Although instruct-tuning does not further amplify this modulation, the pattern reflects a hallmark of successful dialogue dynamics.

Cascade ranking is a widely adopted paradigm in large-scale information retrieval systems for Top-K item selection. However, the Top-K operator is non-differentiable, hindering end-to-end training. Existing methods include Learning-to-Rank approaches (e.g., LambdaLoss), which optimize ranking metrics like NDCG and suffer from objective misalignment, and differentiable sorting-based methods (e.g., ARF, LCRON), which relax permutation matrices for direct Top-K optimization but introduce gradient conflicts through matrix aggregation. A promising alternative is to directly construct a differentiable approximation of the Top-K selection operator, bypassing the use of soft permutation matrices. However, even state-of-the-art differentiable Top-K operator (e.g., LapSum) require $O(n \log n)$ complexity due to their dependence on sorting for solving the threshold. Thus, we propose DFTopK, a novel differentiable Top-K operator achieving optimal $O(n)$ time complexity. By relaxing normalization constraints, DFTopK admits a closed-form solution and avoids sorting. DFTopK also avoids the gradient conflicts inherent in differentiable sorting-based methods. We evaluate DFTopK on both the public benchmark RecFLow and an industrial system. Experimental results show that DFTopK significantly improves training efficiency while achieving superior performance, which enables us to scale up training samples more efficiently. In the online A/B test, DFTopK yielded a +1.77% revenue lift with the same computational budget compared to the baseline. To the best of our knowledge, this work is the first to introduce differentiable Top-K operators into recommendation systems and the first to achieve theoretically optimal linear-time complexity for Top-K selection. We have open-sourced our implementation to facilitate future research in both academia and industry.

LLM-powered agents are both a promising new technology and a source of complexity, where choices about models, tools, and prompting can affect their usefulness. While numerous benchmarks measure agent accuracy across domains, they mostly assume full automation, failing to represent the collaborative nature of real-world use cases. In this paper, we make two major steps towards the rigorous assessment of human-agent interactions. First, we propose PULSE, a framework for more efficient human-centric evaluation of agent designs, which comprises collecting user feedback, training an ML model to predict user satisfaction, and computing results by combining human satisfaction ratings with model-generated pseudo-labels. Second, we deploy the framework on a large-scale web platform built around the open-source software agent OpenHands, collecting in-the-wild usage data across over 15k users. We conduct case studies around how three agent design decisions -- choice of LLM backbone, planning strategy, and memory mechanisms -- impact developer satisfaction rates, yielding practical insights for software agent design. We also show how our framework can lead to more robust conclusions about agent design, reducing confidence intervals by 40% compared to a standard A/B test. Finally, we find substantial discrepancies between in-the-wild results and benchmark performance (e.g., the anti-correlation between results comparing claude-sonnet-4 and gpt-5), underscoring the limitations of benchmark-driven evaluation. Our findings provide guidance for evaluations of LLM agents with humans and identify opportunities for better agent designs.

Reward model (RM) plays a pivotal role in aligning large language model (LLM) with human preferences. As real-world applications increasingly involve long history trajectories, e.g., LLM agent, it becomes indispensable to evaluate whether a model's responses are not only high-quality but also grounded in and consistent with the provided context. Yet, current RMs remain confined to short-context settings and primarily focus on response-level attributes (e.g., safety or helpfulness), while largely neglecting the critical dimension of long context-response consistency. In this work, we introduce Long-RewardBench, a benchmark specifically designed for long-context RM evaluation, featuring both Pairwise Comparison and Best-of-N tasks. Our preliminary study reveals that even state-of-the-art generative RMs exhibit significant fragility in long-context scenarios, failing to maintain context-aware preference judgments. Motivated by the analysis of failure patterns observed in model outputs, we propose a general multi-stage training strategy that effectively scales arbitrary models into robust Long-context RMs (LongRMs). Experiments show that our approach not only substantially improves performance on long-context evaluation but also preserves strong short-context capability. Notably, our 8B LongRM outperforms much larger 70B-scale baselines and matches the performance of the proprietary Gemini 2.5 Pro model.

Language models are increasingly applied to biological sequences like proteins and mRNA, yet their default Euclidean geometry may mismatch the hierarchical structures inherent to biological data. While hyperbolic geometry provides a better alternative for accommodating hierarchical data, it has yet to find a way into language modeling for mRNA sequences. In this work, we introduce HyperHELM, a framework that implements masked language model pre-training in hyperbolic space for mRNA sequences. Using a hybrid design with hyperbolic layers atop Euclidean backbone, HyperHELM aligns learned representations with the biological hierarchy defined by the relationship between mRNA and amino acids. Across multiple multi-species datasets, it outperforms Euclidean baselines on 9 out of 10 tasks involving property prediction, with 10% improvement on average, and excels in out-of-distribution generalization to long and low-GC content sequences; for antibody region annotation, it surpasses hierarchy-aware Euclidean models by 3% in annotation accuracy. Our results highlight hyperbolic geometry as an effective inductive bias for hierarchical language modeling of mRNA sequences. Language models have been increasingly applied to biological sequence data, fueled by the growth of large-scale omics datasets (Lin et al., 2023; Celaj et al., 2023; Brixi et al., 2025). The biological sequences, however, are structured differently from natural language, particularly in their hierarchical organization, where nucleotides or amino acids form motifs that can be nested within larger functional groups (Buhr et al., 2016). In this work, we take the rapidly expanding therapeutic domain of RNA, where the codon-amino acid hierarchy plays a key role in determining the biophysical properties of mRNA sequences and their expressed proteins (Clancy & Brown, 2008), and we focus on encoding this hierarchy directly into the representation space of a bio-language model by leveraging hyperbolic geometry.

Drug target interaction (DTI) prediction is a cornerstone of computational drug discovery, enabling rational design, repurposing, and mechanistic insights. While deep learning has advanced DTI modeling, existing approaches primarily rely on SMILES protein pairs and fail to exploit the rich multimodal information available for small molecules and proteins. We introduce GRAMDTI, a pretraining framework that integrates multimodal molecular and protein inputs into unified representations. GRAMDTI extends volume based contrastive learning to four modalities, capturing higher-order semantic alignment beyond conventional pairwise approaches. To handle modality informativeness, we propose adaptive modality dropout, dynamically regulating each modality's contribution during pre-training. Additionally, IC50 activity measurements, when available, are incorporated as weak supervision to ground representations in biologically meaningful interaction strengths. Experiments on four publicly available datasets demonstrate that GRAMDTI consistently outperforms state of the art baselines. Our results highlight the benefits of higher order multimodal alignment, adaptive modality utilization, and auxiliary supervision for robust and generalizable DTI prediction.

Although transformer architectures have achieved state-of-the-art performance across diverse domains, their quadratic computational complexity with respect to sequence length remains a significant bottleneck, particularly for latency-sensitive long-context applications. While recent linear-complexity alternatives are increasingly powerful, effectively training them from scratch is still resource-intensive. To overcome these limitations, we propose LAWCAT (Linear Attention with Convolution Across Time), a novel linearization framework designed to efficiently transfer the capabilities of pre-trained transformers into a performant linear attention architecture. LAWCAT integrates causal Conv1D layers to enhance local dependency modeling and employs normalized gated linear attention to improve generalization across varying context lengths. Our comprehensive evaluations demonstrate that, distilling Mistral-7B with only 1K-length sequences yields over 90\% passkey retrieval accuracy up to 22K tokens, significantly extending its effective context window. Similarly, Llama3.2-1B LAWCAT variant achieves competitive performance on S-NIAH 1\&2\&3 tasks (1K-8K context length) and BABILong benchmark (QA2\&QA3, 0K-16K context length), requiring less than 0.1\% pre-training tokens compared with pre-training models. Furthermore, LAWCAT exhibits faster prefill speeds than FlashAttention-2 for sequences exceeding 8K tokens. LAWCAT thus provides an efficient pathway to high-performance, long-context linear models suitable for edge deployment, reducing reliance on extensive long-sequence training data and computational resources. Code is released at: https://github.com/zeyuliu1037/LAWCAT