Clinical trial amendments frequently introduce delays, increased costs, and administrative burden, with eligibility criteria being the most commonly amended component. We introduce \textit{eligibility criteria amendment prediction}, a novel NLP task that aims to forecast whether the eligibility criteria of an initial trial protocol will undergo future amendments. To support this task, we release $\texttt{AMEND++}$, a benchmark suite comprising two datasets: $\texttt{AMEND}$, which captures eligibility-criteria version histories and amendment labels from public clinical trials, and $\verb|AMEND_LLM|$, a refined subset curated using an LLM-based denoising pipeline to isolate substantive changes. We further propose $\textit{Change-Aware Masked Language Modeling}$ (CAMLM), a revision-aware pretraining strategy that leverages historical edits to learn amendment-sensitive representations. Experiments across diverse baselines show that CAMLM consistently improves amendment prediction, enabling more robust and cost-effective clinical trial design.
We study group decision-making in artificial societies where the rules of play are themselves subject to collective amendment. Using the self-amending game Nomic, we compare multiple scales across two LLM families and find that collective adaptation does not improve monotonically with model size. Instead, both families exhibit a narrow mid-scale regime that supports sustained rule adoption, diverse amendments, and balanced consensus. Smaller models tend to remain rule-inert, whereas larger models often converge on restrictive voting patterns, and heterogeneous mixed-size groups collapse into veto-driven gridlock. These cross-scale contrasts persist under temperature perturbations and under a shift from unanimity to majority voting, although latent-state structure varies by family and scale. Hidden-state divergence alone does not explain collective performance: high representational divergence can coincide with poor behavioural outcomes. Linear probes reveal regime-selective coupling between latent vote-predictive signals and collective behaviour, but decodability is necessary rather than sufficient for adaptive play. Overall, the recurring regularity is non-monotonicity, not the pa
Long-context decoding in LLMs is IO-bound: each token re-reads an ever-growing KV cache. Prior accelerations cut bytes via compression, which lowers fidelity, or selection/eviction, which restricts what remains accessible, and both can degrade delayed recall and long-form generation. We introduce MAC-Attention, a fidelity- and access-preserving alternative that accelerates decoding by reusing prior attention computations for semantically similar recent queries. It starts with a match stage that performs pre-RoPE L2 matching over a short local window; an amend stage rectifies the reused attention by recomputing a small band near the match boundary; and a complete stage fuses the rectified results with fresh attention computed on the KV tail through a numerically stable merge. On a match hit, the compute and bandwidth complexity is constant regardless of context length. The method is model-agnostic and composes with IO-aware kernels, paged-KV managers, and MQA/GQA. Across LongBench v2 (120K), RULER (120K), and LongGenBench (16K continuous generation), compared to the latest FlashInfer library, MAC-Attention reduces KV accesses by up to 99%, cuts token generation latency by over 60% a
It has been observed that deep neural networks (DNNs) often use both genuine as well as spurious features. In this work, we propose "Amending Inherent Interpretability via Self-Supervised Masking" (AIM), a simple yet interestingly effective method that promotes the network's utilization of genuine features over spurious alternatives without requiring additional annotations. In particular, AIM uses features at multiple encoding stages to guide a self-supervised, sample-specific feature-masking process. As a result, AIM enables the training of well-performing and inherently interpretable models that faithfully summarize the decision process. We validate AIM across a diverse range of challenging datasets that test both out-of-distribution generalization and fine-grained visual understanding. These include general-purpose classification benchmarks such as ImageNet100, HardImageNet, and ImageWoof, as well as fine-grained classification datasets such as Waterbirds, TravelingBirds, and CUB-200. AIM demonstrates significant dual benefits: interpretability improvements, as measured by the Energy Pointing Game (EPG) score, and accuracy gains over strong baselines. These consistent gains acro
We introduce a formal active learning methodology for guiding the placement of Lagrangian observers to infer time-dependent vector fields -- a key task in oceanography, marine science, and ocean engineering -- using a physics-informed spatio-temporal Gaussian process surrogate model. The majority of existing placement campaigns either follow standard `space-filling' designs or relatively ad-hoc expert opinions. A key challenge to applying principled active learning in this setting is that Lagrangian observers are continuously advected through the vector field, so they make measurements at different locations and times. It is, therefore, important to consider the likely future trajectories of placed observers to account for the utility of candidate placement locations. To this end, we present BALLAST: Bayesian Active Learning with Look-ahead Amendment for Sea-drifter Trajectories. We observe noticeable benefits of BALLAST-aided sequential observer placement strategies on both synthetic and high-fidelity ocean current models. In addition, we developed a novel GP inference method -- the Vanilla SPDE Exchange (VaSE) -- to boost the GP posterior sampling efficiency, which is also of ind
We show that the amended monadic Grzegorczyk logic $\mathsf{M^+Grz}$ is the largest modal companion of the amended monadic intuitionistic logic $\mathsf{M^+IPC}$. Thus, unlike the monadic intuitionistic logic $\mathsf{MIPC}$, Esakia's theorem does extend to $\mathsf{M^+IPC}$.
This paper challenges the assumption that courts should grant First Amendment protections to outputs from large generative AI models, such as GPT-4 and Gemini. We argue that because these models lack intentionality, their outputs do not constitute speech as understood in the context of established legal precedent, so there can be no speech to protect. Furthermore, if the model outputs are not speech, users cannot claim a First Amendment speech right to receive the outputs. We also argue that extending First Amendment rights to AI models would not serve the fundamental purposes of free speech, such as promoting a marketplace of ideas, facilitating self-governance, or fostering self-expression. In fact, granting First Amendment protections to AI models would be detrimental to society because it would hinder the government's ability to regulate these powerful technologies effectively, potentially leading to the unchecked spread of misinformation and other harms.
In order to establish the thermodynamic stability of a system, knowledge of its Gibbs free energy is essential. Most often, the Gibbs free energy is predicted within the CALPHAD framework using models employing thermodynamic properties, such as the mixing enthalpy, heat capacity, and activity coefficients. Here, we present a deep-learning approach capable of predicting the mixing enthalpy of liquid phases of binary systems that were not present in the training dataset. Therefore, our model allows for a system-informed enhancement of the thermodynamic description to unknown binary systems based on information present in the available thermodynamic assessment. Thereby, significant experimental efforts in assessing new systems can be spared. We use an open database for steels containing 91 binary systems to generate our initial training (and validation) and amend it with several direct experimental reports. The model is thoroughly tested using different strategies, including a test of its predictive capabilities. The model shows excellent predictive capabilities outside of the training dataset as soon as some data containing species of the predicted system is included in the training
The successive and the amendment procedures have been widely employed in parliamentary and legislative decision making and have undergone extensive study in the literature from various perspectives. However, investigating them through the lens of computational complexity theory has not been as thoroughly conducted as for many other prevalent voting procedures heretofore. To the best of our knowledge, there is only one paper which explores the complexity of several strategic voting problems under these two procedures, prior to our current work. To provide a better understanding of to what extent the two procedures resist strategic behavior, we study the parameterized complexity of constructive/destructive control by adding/deleting voters/candidates for both procedures. To enhance the generalizability of our results, we also examine a more generalized form of the amendment procedure. Our exploration yields a comprehensive (parameterized) complexity landscape of these problems with respect to numerous parameters.
Quadrotor control policies can be trained with high performance using the exact gradients of the rewards to directly optimize policy parameters via backpropagation-through-time (BPTT). However, designing a fully differentiable reward architecture is often challenging. Partially differentiable rewards will result in biased gradient propagation that degrades training performance. To overcome this limitation, we propose Amended Backpropagation-through-Time (ABPT), a novel approach that mitigates gradient bias while preserving the training efficiency of BPTT. ABPT combines 0-step and N-step returns, effectively reducing the bias by leveraging value gradients from the learned Q-value function. Additionally, it adopts entropy regularization and state initialization mechanisms to encourage exploration during training. We evaluate ABPT on four representative quadrotor flight tasks \li{in both real world and simulation}. Experimental results demonstrate that ABPT converges significantly faster and achieves higher ultimate rewards than existing learning algorithms, particularly in tasks involving partially differentiable rewards. The code will be released at http://github.com/Fanxing-LI/ABPT
It is known that the usual form of the Ehrenfest theorem (ET), which couples the motion of the center of mass (COM) of the one-dimensional (1D) wave function to the respective classical equation of motion, is not valid in the case of the potential box, confined by the zero boundary conditions. A modified form of the ET was proposed for this case, which includes an effective force originating from the interaction of the 1D quantum particle with the box edges. In this work, we derive an amended ET for the Gross-Pitaevskii equation (GPE), which includes the cubic nonlinear term, as well as for the 2D square-shaped potential box. In the latter case, we derive an amended COM equation of motion with an effective force exerted by the edges of the rectangular box, while the nonlinear term makes no direct contribution to the 1D and 2D versions of the ET. Nonetheless, the nonlinearity affects the amended ET through the edge-generated force. As a result, the nonlinearity of the underlying GPE can make the COM motion in the potential box irregular. The validity of the amended ET for the 1D and 2D GPEs with the respective potential boxes is confirmed by the comparison of numerical simulations o
The worm gearbox is a high-speed transmission system that plays a vital role in various industries. Therefore it becomes necessary to develop a robust fault diagnosis scheme for worm gearbox. Due to advancements in sensor technology, researchers from academia and industries prefer deep learning models for fault diagnosis purposes. The optimal selection of hyperparameters (HPs) of deep learning models plays a significant role in stable performance. Existing methods mainly focused on manual tunning of these parameters, which is a troublesome process and sometimes leads to inaccurate results. Thus, exploring more sophisticated methods to optimize the HPs automatically is important. In this work, a novel optimization, i.e. amended gorilla troop optimization (AGTO), has been proposed to make the convolutional neural network (CNN) adaptive for extracting the features to identify the worm gearbox defects. Initially, the vibration and acoustic signals are converted into 2D images by the Morlet wavelet function. Then, the initial model of CNN is developed by setting hyperparameters. Further, the search space of each Hp is identified and optimized by the developed AGTO algorithm. The classif
The reasoning abilities are one of the most enigmatic and captivating aspects of large language models (LLMs). Numerous studies are dedicated to exploring and expanding the boundaries of this reasoning capability. However, tasks that embody both reasoning and recall characteristics are often overlooked. In this paper, we introduce such a novel task, code reasoning, to provide a new perspective for the reasoning abilities of LLMs. We summarize three meta-benchmarks based on established forms of logical reasoning, and instantiate these into eight specific benchmark tasks. Our testing on these benchmarks reveals that LLMs continue to struggle with identifying satisfactory reasoning pathways. Additionally, we present a new pathway exploration pipeline inspired by human intricate problem-solving methods. This Reflective Hypothesis Decomposition and Amendment (RHDA) pipeline consists of the following iterative steps: (1) Proposing potential hypotheses based on observations and decomposing them; (2) Utilizing tools to validate hypotheses and reflection outcomes; (3) Revising hypothesis in light of observations. Our approach effectively mitigates logical chain collapses arising from forget
Quantum mechanics of unitary systems is considered in quasi-Hermitian representation. In this framework the concept of perturbation is found counterintuitive, for three reasons. The first one is that in this formalism we are allowed to change the physical Hilbert-space norm. Thus, in a preselected Hamiltonian $H(λ)=H_0+λ\,H_1$ the size (and, hence, influence) of the perturbation cannot always be kept under a reliable control. Often, an enhanced sensitivity to perturbations is observed, for this reason, in open quantum systems. Second, even when we consider just a closed quantum system in which the influence of $H_1 eq H_1^\dagger$ is guaranteed to be small, the correct probabilistic interpretation of the system remains ambiguous, mainly due to the non-uniqueness of the physical Hilbert-space inner-product metric~$Θ$. Third, even if we decide to ignore the ambiguity and if we pick up just any one of the eligible metrics (which reduces the scope of the theory of course), such a choice would still vary with $λ$. In our paper it is shown that all of these three obstacles can be circumvented via just a mild amendment of the Rayleigh-Schrödinger perturbation-expansion approach. The flexi
In this paper, we develop a systematic deep learning approach to solve two-dimensional (2D) stationary quantum droplets (QDs) and investigate their wave propagation in the 2D amended Gross-Pitaevskii equation with Lee-Huang-Yang correction and two kinds of potentials. Firstly, we use the initial-value iterative neural network (IINN) algorithm for 2D stationary quantum droplets of stationary equations. Then the learned stationary QDs are used as the initial value conditions for physics-informed neural networks (PINNs) to explore their evolutions in the some space-time region. Especially, we consider two types of potentials, one is the 2D quadruple-well Gaussian potential and the other is the PT-symmetric HO-Gaussian potential, which lead to spontaneous symmetry breaking and the generation of multi-component QDs. The used deep learning method can also be applied to study wave propagations of other nonlinear physical models.
An in-process coherent imaging diagnostic has been developed to real-time measure the hole depth during air-film hole drilling by a femtosecond laser. A super-luminescent diode with a wavelength of 830~13 nm is chosen as the coherent light source which determines a depth resolution of 12 μm. The drilled hole is coupled as a part of the sample arm and the depth variation can be extracted from the length variation of the optical path. Interference is realized in the detection part and a code has been written to discriminate the interference fringes. Density of plasma in the hole is diagnosed to evaluate its amendment to the optical path length and the depth measurement error induced by plasma is non-ignorable when drilling deep holes.
Named Entity Recognition (NER) is one of the essential applications of Natural Language Processing (NLP). It is also an instrument that plays a significant role in many other NLP applications, such as Machine Translation (MT), Information Retrieval (IR), and Part of Speech Tagging (POST). Kurdish is an under-resourced language from the NLP perspective. Particularly, in all the categories, the lack of NER resources hinders other aspects of Kurdish processing. In this work, we present a data set that covers several categories of NEs in Kurdish (Sorani). The dataset is a significant amendment to a previously developed dataset in the Kurdish BLARK (Basic Language Resource Kit). It covers 11 categories and 33261 entries in total. The dataset is publicly available for non-commercial use under CC BY-NC-SA 4.0 license at https://kurdishblark.github.io/.
Adversarial attack is commonly regarded as a huge threat to neural networks because of misleading behavior. This paper presents an opposite perspective: adversarial attacks can be harnessed to improve neural models if amended correctly. Unlike traditional adversarial defense or adversarial training schemes that aim to improve the adversarial robustness, the proposed adversarial amendment (AdvAmd) method aims to improve the original accuracy level of neural models on benign samples. We thoroughly analyze the distribution mismatch between the benign and adversarial samples. This distribution mismatch and the mutual learning mechanism with the same learning ratio applied in prior art defense strategies is the main cause leading the accuracy degradation for benign samples. The proposed AdvAmd is demonstrated to steadily heal the accuracy degradation and even leads to a certain accuracy boost of common neural models on benign classification, object detection, and segmentation tasks. The efficacy of the AdvAmd is contributed by three key components: mediate samples (to reduce the influence of distribution mismatch with a fine-grained amendment), auxiliary batch norm (to solve the mutual
The synthetic control estimator (Abadie et al., 2010) is asymptotically unbiased assuming that the outcome is a linear function of the underlying predictors and that the treated unit can be well approximated by the synthetic control before the treatment. When the outcome is nonlinear, the bias of the synthetic control estimator can be severe. In this paper, we provide conditions for the synthetic control estimator to be asymptotically unbiased when the outcome is nonlinear, and propose a flexible and data-driven method to choose the synthetic control weights. Monte Carlo simulations show that compared with the competing methods, the nonlinear synthetic control method has similar or better performance when the outcome is linear, and better performance when the outcome is nonlinear, and that the confidence intervals have good coverage probabilities across settings. In the empirical application, we illustrate the method by estimating the impact of the 2019 anti-extradition law amendments bill protests on Hong Kong's economy, and find that the year-long protests reduced real GDP per capita in Hong Kong by 11.27% in the first quarter of 2020, which was larger in magnitude than the econo