Oligonucleotide therapeutics offer great potential to address previously undruggable targets and enable personalized medicine. However, their progress is often hindered by insufficient safety and efficacy profiles. Predictive modeling and machine learning could significantly accelerate oligonucleotide drug discovery by identifying suboptimal compounds early on, but their application in this area lags behind other modalities. A key obstacle to the adoption of machine learning in the field is the scarcity of readily accessible and standardized datasets for model development, as data are often scattered across diverse experiments with inconsistent molecular representations. To overcome this challenge, we introduce OligoGym, a curated collection of standardized, machine learning-ready datasets encompassing various oligonucleotide therapeutic modalities and endpoints. We used OligoGym to benchmark diverse classical and deep learning methods, establishing performance baselines for each dataset across different featurization techniques, model configurations, and splitting strategies. Our work represents a crucial first step in creating a more unified framework for oligonucleotide therapeutic dataset generation and model training.

Mechanistic interpretability (MI) seeks to uncover how language models (LMs) implement specific behaviors, yet measuring progress in MI remains challenging. The recently released Mechanistic Interpretability Benchmark (MIB; Mueller et al., 2025) provides a standardized framework for evaluating circuit and causal variable localization. Building on this foundation, the BlackboxNLP 2025 Shared Task extends MIB into a community-wide reproducible comparison of MI techniques. The shared task features two tracks: circuit localization, which assesses methods that identify causally influential components and interactions driving model behavior, and causal variable localization, which evaluates approaches that map activations into interpretable features. With three teams spanning eight different methods, participants achieved notable gains in circuit localization using ensemble and regularization strategies for circuit discovery. With one team spanning two methods, participants achieved significant gains in causal variable localization using low-dimensional and non-linear projections to featurize activation vectors. The MIB leaderboard remains open; we encourage continued work in this standard evaluation framework to measure progress in MI research going forward.

To avoid the featuri zation bottleneck of traditional descriptors, we also leverage GNNs to generate fast, latent-space approximations of MatMiner (ℓ-MM) and Orbital Field Matrix (ℓ-OFM) features. Finally, we augment this feature set with new descriptors derived via symbolic regression. This multifac eted strategy aims to create a more robust, accurate, and versatile featurizer that capitalizes on the distinct strengths of each approach to be useful for a wider range of dataset sizes. To simplify the generation of all those features, a package was developed named MatterVial standing for MATerials fea T uR e E xtraction Via I nterpretable Artificial L earning, which, besides producing all latent-space features from the GNN models, aids i n obtaining the interpretable chemical descriptors that correlate to these high-level features. This is achieved through techniques such as SHapley Additive exPlanations (SHAP) analysi s in surrogate models and symbolic regression via Sure Independence Screening and Sparsifying Operator (SISSO) to obtain an approximate formula from the most important features. Our re sults demonstrate an overall improvement in all analyzed datasets compare d with the baseline MatMiner featurizer. In addition, it surpassed the performance of the individua l GNN models in several cases, indicating that the combination of traditional and l atent-space features leads to a more robust generalization.

The need to develop a general framework for architecture analysis is becoming increasingly important, given the expanding design space of sequence models. To this end, we draw insights from classical signal processing and control theory, to develop a quantitative measure of \textit{memory utilization}: the internal mechanisms through which a model stores past information to produce future outputs. This metric, which we call \textbf{\textit{effective state-size}} (ESS), is tailored to the fundamental class of systems with \textit{input-invariant} and \textit{input-varying linear operators}, encompassing a variety of computational units such as variants of attention, convolutions, and recurrences. Unlike prior work on memory utilization, which either relies on raw operator visualizations (e.g. attention maps), or simply the total \textit{memory capacity} (i.e. cache size) of a model, our metrics provide highly interpretable and actionable measurements. In particular, we show how ESS can be leveraged to improve initialization strategies, inform novel regularizers and advance the performance-efficiency frontier through model distillation. Furthermore, we demonstrate that the effect of context delimiters (such as end-of-speech tokens) on ESS highlights cross-architectural differences in how large language models utilize their available memory to recall information. Overall, we find that ESS provides valuable insights into the dynamics that dictate memory utilization, enabling the design of more efficient and effective sequence models.

Iterative improvement of model architectures is fundamental to deep learning: Transformers first enabled scaling, and recent advances in model hybridization have pushed the quality-efficiency frontier. However, optimizing architectures remains challenging and expensive. Current automated or manual approaches fall short, largely due to limited progress in the design of search spaces and due to the simplicity of resulting patterns and heuristics. In this work, we propose a new approach for the synthesis of tailored architectures (STAR). Our approach combines a novel search space based on the theory of linear input-varying systems, supporting a hierarchical numerical encoding into architecture genomes. STAR genomes are automatically refined and recombined with gradient-free, evolutionary algorithms to optimize for multiple model quality and efficiency metrics. Using STAR, we optimize large populations of new architectures, leveraging diverse computational units and interconnection patterns, improving over highly-optimized Transformers and striped hybrid models on the frontier of quality, parameter size, and inference cache for autoregressive language modeling.

The use of machine learning methods in high energy physics typically relies on large volumes of precise simulation for training. As machine learning models become more complex they can become increasingly sensitive to differences between this simulation and the real data collected by experiments. We present a generic methodology based on contrastive learning which is able to greatly mitigate this negative effect. Crucially, the method does not require prior knowledge of the specifics of the mismodeling. While we demonstrate the efficacy of this technique using the task of jet-tagging at the Large Hadron Collider, it is applicable to a wide array of different tasks both in and out of the field of high energy physics.

Generalization and optimization guarantees on the population loss in machine learning often rely on uniform convergence based analysis, typically based on the Rademacher complexity of the predictors. The rich representation power of modern models has led to concerns about this approach. In this paper, we present generalization and optimization guarantees in terms of the complexity of the gradients, as measured by the Loss Gradient Gaussian Width (LGGW). First, we introduce generalization guarantees directly in terms of the LGGW under a flexible gradient domination condition, which we demonstrate to hold empirically for deep models. Second, we show that sample reuse in finite sum (stochastic) optimization does not make the empirical gradient deviate from the population gradient as long as the LGGW is small. Third, focusing on deep networks, we present results showing how to bound their LGGW under mild assumptions. In particular, we show that their LGGW can be bounded (a) by the $L_2$-norm of the loss Hessian eigenvalues, which has been empirically shown to be $\tilde{O}(1)$ for commonly used deep models; and (b) in terms of the Gaussian width of the featurizer, i.e., the output of the last-but-one layer. To our knowledge, our generalization and optimization guarantees in terms of LGGW are the first results of its kind, avoid the pitfalls of predictor Rademacher complexity based analysis, and hold considerable promise towards quantitatively tight bounds for deep models.

Machine learning algorithms minimizing the average training loss usually suffer from poor generalization performance due to the greedy exploitation of correlations among the training data, which are not stable under distributional shifts. It inspires various works for domain generalization (DG), where a series of methods, such as Causal Matching and FISH, work by pairwise domain operations. They would need $O(n^2)$ pairwise domain operations with $n$ domains, where each one is often highly expensive. Moreover, while a common objective in the DG literature is to learn invariant representations against domain-induced spurious correlations, we highlight the importance of mitigating spurious correlations caused by objects. Based on the observation that diversity helps mitigate spurious correlations, we propose a Diversity boosted twO-level saMplIng framework (DOMI) utilizing Determinantal Point Processes (DPPs) to efficiently sample the most informative ones among large number of domains. We show that DOMI helps train robust models against spurious correlations from both domain-side and object-side, substantially enhancing the performance of the backbone DG algorithms on rotated MNIST, rotated Fashion MNIST, and iwildcam datasets.

In the history of conversational AI agents, ELIZA [2], [3], one In this era of artificial intelligence (AI), chatbots are becoming of the first rule-based chatbots, took it upon itself to pass the more and more popular every day for their versatility, easy famous Turing Test and pioneer the path of guided computer accessibility, personalizing features, and, more importantly, responses. Even though it failed to pass the test completely, it their ability to generate automated responses. Specifically for surely did not come short in paving the way for other artificial these purposes, we now see an uprise of chatbots everywhere - chatbots, which ranged from responding emotionally (PARRY) from personal to organizational, to business websites or other [2]-[4] to simply having fun conversations by running pattern online platforms, for which it can be trained on suitable data matching (Jabberwacky) [2]. Later, this field got more to make it, in a broader sense, a virtual assistant representative matured with the inception of AI-powered chatbots, namely of the said entities. Dr. Sbaitso [3] and A.L.I.C.E (Artificial Linguistic Internet In the light of this newly emerging scope, we explore the Computer Entity) [2], [4]- which was able to mimic humans possibilities of how these conversational AI agents can be when chatting online or answering questions. From there, integrated properly and thus be an immensely useful tool to it was not long before Smarterchild, Siri, Google Assistant, maintain business activities. To better understand the concurrent and other personalized assistant-like chatbots or conversational chatbots and find possible modifications in them and AI agents came into existence. With conversational AI, now, for further and more customized improvements, we choose a anyone can build, integrate, and use message-based or speechbased trendy chatbot platform Rasa as our study subject.

Reinforcement learning (RL) has recently shown impressive performance in complex game AI and robotics tasks. To a large extent, this is thanks to the availability of simulated environments such as OpenAI Gym, Atari Learning Environment, or Malmo which allow agents to learn complex tasks through interaction with virtual environments. While RL is also increasingly applied to natural language processing (NLP), there are no simulated textual environments available for researchers to apply and consistently benchmark RL on NLP tasks. With the work reported here, we therefore release NLPGym, an open-source Python toolkit that provides interactive textual environments for standard NLP tasks such as sequence tagging, multi-label classification, and question answering. We also present experimental results for 6 tasks using different RL algorithms which serve as baselines for further research. The toolkit is published at https://github.com/rajcscw/nlp-gym