arXiv subjects

David Harel

Publications and source records attributed to David Harel.

At least 19 recordsLinked to original sources

arXiv2026

Rethinking Higher Education: From Fixed Curricula to Learnity Graphs

Higher education stands at a turning point. In an era where knowledge is increasingly accessible and which is, more often than not, mediated by advanced Artificial Intelligence (AI), the value of traditional curricula models warrants reconsideration. This does not imply that one should replace thorough academic studies. Universities remain essential in providing foundational knowledge, theoretical depth and conceptual grounding. The challenge is to extend these educational facets with learning environments that foster creativity, interdisciplinary integration, hands-on experience, and especially long-term development. In this paper, we introduce a lifelong learning framework that integrates academic, professional, and personal learning, centered on a new concept that we term learnity graphs, a structured representation of learning as interconnected units of knowledge, skills, experience, and actual artifacts, coupled with a method for presenting, and leveraging it.

cs.CY↗

arXiv2026

On Training in Imagination

State-of-the-art model-based reinforcement learning methods train policies on imagined rollouts. These rollouts are trajectories generated by a learned dynamics model and are scored by a learned reward model, but without querying the true environment during policy updates. We study this training paradigm by quantifying how errors in learned dynamics and reward models affect returns and policy optimization. First, we extend the analysis of Asadi et al. (2018) to MDPs with learned reward models, and derive the optimal sample allocation--the ratio of dynamics samples to reward samples that minimizes a bound on return error under power-law scaling assumptions. We identify lower Lipschitz constants of the learned dynamics, reward, and policy as a representation desideratum that tightens this bound, and we connect this perspective to the temporal-straightening objective of Wang et al. (2026). Second, we examine how policy optimization with REINFORCE tolerates noisy rewards, which are often cheaper to obtain. We show that zero-mean reward noise leaves the gradient estimator unbiased and adds at most a variance term that decreases with the number of rollouts. This introduces a practical tradeoff: given a fixed budget, should one buy more rollouts with cheaper but noisier rewards, or fewer rollouts with more expensive but less noisy rewards? We reduce this choice to a one-dimensional optimization problem and characterize the optimum.

cs.LG↗

arXiv2025

A Specification's Realm: Characterizing the Knowledge Required for Executing a Given Algorithm Specification

An algorithm specification in natural language or pseudocode is expected to be clear and explicit enough to enable mechanical execution. In this position paper we contribute an initial characterization of the knowledge that an executing agent, human or machine, should possess in order to be able to carry out the instructions of a given algorithm specification as a stand-alone entity, independent of any system implementation. We argue that, for that algorithm specification, such prerequisite knowledge, whether unique or shared with other specifications, can be summarized in a document of practical size. We term this document the realm of the algorithm specification. The generation of such a realm is itself a systematic analytical process, significant parts of which can be automated with the help of large language models and the reuse of existing documents. The algorithm-specification's realm would consist of specification language syntax and semantics, domain knowledge restricted to the referenced entities, inter-entity relationships, relevant underlying cause-and-effect rules, and detailed instructions and means for carrying out certain operations. Such characterization of the realm can contribute to methodological implementation of the algorithm specification in diverse systems and to its formalization for mechanical verification. The paper also touches upon the question of assessing execution faithfulness, which is distinct from correctness: in the absence of a reference interpretation of natural language or pseudocode specification with a given vocabulary, how can we determine if an observed agent's execution indeed complies with the input specification.

cs.PL↗

arXiv2025

Meta-autoencoders: An approach to discovery and representation of relationships between dynamically evolving classes

An autoencoder (AE) is a neural network that, using self-supervised training, learns a succinct parameterized representation, and a corresponding encoding and decoding process, for all instances in a given class. Here, we introduce the concept of a meta-autoencoder (MAE): an AE for a collection of autoencoders. Given a family of classes that differ from each other by the values of some parameters, and a trained AE for each class, an MAE for the family is a neural net that has learned a compact representation and associated encoder and decoder for the class-specific AEs. One application of this general concept is in research and modeling of natural evolution -- capturing the defining and the distinguishing properties across multiple species that are dynamically evolving from each other and from common ancestors. In this interim report we provide a constructive definition of MAEs, initial examples, and the motivating research directions in machine learning and biology.

cs.LG↗

arXiv2025

Out-of-Vocabulary Sampling Boosts Speculative Decoding

Speculative decoding relies on fast and accurate drafters. Recent state-of-the-art language models employ larger and larger vocabularies, which significantly slows down drafters. One promising approach to boost the efficiency of speculative decoding is to use drafters with smaller vocabularies. However, existing sampling methods cannot draw out-of-vocabulary tokens, creating a tradeoff between drafters' vocabulary size and acceptance rates. This paper introduces Redistributing Drafter Kernels (RDK), the first out-of-vocabulary sampler that effectively recovers acceptance rates by virtually restoring pruned target tokens. RDK leverages token-affinity priors to reallocate drafter mass towards high-overlap regions. We prove mathematically that RDK can achieve higher acceptance rates than vanilla and state-of-the-art samplers. We provide an efficient first-order approximation of RDK and prove that it reduces redistribution times from $O(N^2)$ to $O(N)$, enabling lightweight implementations for large vocabularies. Our experiments demonstrate that this linear-time RDK significantly boosts acceptance rates even after extreme pruning (removing more than 75% of the drafter's vocabulary), where existing samplers fail. RDK opens the door to extremely pruned drafters, which were previously impractical.

cs.LG↗

arXiv2025

Accelerating LLM Inference with Lossless Speculative Decoding Algorithms for Heterogeneous Vocabularies

Accelerating the inference of large language models (LLMs) is a critical challenge in generative AI. Speculative decoding (SD) methods offer substantial efficiency gains by generating multiple tokens using a single target forward pass. However, existing SD approaches require the drafter and target models to share the same vocabulary, thus limiting the pool of possible drafters, often necessitating the training of a drafter from scratch. We present three new SD methods that remove this shared-vocabulary constraint. All three methods preserve the target distribution (i.e., they are lossless) and work with off-the-shelf models without requiring additional training or modifications. Empirically, on summarization, programming, and long-context tasks, our algorithms demonstrate significant speedups of up to 2.8x over standard autoregressive decoding. By enabling any off-the-shelf model to serve as a drafter and requiring no retraining, this work substantially broadens the applicability of the SD framework in practice.

cs.CL↗

arXiv2025

Data and System Perspectives of Sustainable Artificial Intelligence

Sustainable AI is a subfield of AI for concerning developing and using AI systems in ways of aiming to reduce environmental impact and achieve sustainability. Sustainable AI is increasingly important given that training of and inference with AI models such as large langrage models are consuming a large amount of computing power. In this article, we discuss current issues, opportunities and example solutions for addressing these issues, and future challenges to tackle, from the data and system perspectives, related to data acquisition, data processing, and AI model training and inference.

cs.AI↗

arXiv2025

An Infrastructure Software Perspective Toward Computation Offloading between Executable Specifications and Foundation Models

Foundation Models (FMs) have become essential components in modern software systems, excelling in tasks such as pattern recognition and unstructured data processing. However, their capabilities are complemented by the precision, verifiability, and deterministic nature of executable specifications, such as symbolic programs. This paper explores a new perspective on computation offloading, proposing a framework that strategically distributes computational tasks between FMs and executable specifications based on their respective strengths. We discuss the potential design of an infrastructure software framework to enable this offloading, focusing on key mechanisms such as task decomposition, resource allocation, and adaptive optimization. Furthermore, we identify critical technical challenges, including semantic-gap resolution, reliability, and scalability, that must be addressed to realize this approach. By leveraging the complementary strengths of FMs and symbolic programs, this perspective lays the groundwork for advancing hybrid software systems that are both efficient and reliable.

cs.SE↗

arXiv2025

Beyond Pass or Fail: Multi-Dimensional Benchmarking of Foundation Models for Goal-based Mobile UI Navigation

Recent advances of foundation models (FMs) have made navigating mobile applications (apps) based on high-level goal instructions within reach, with significant industrial applications such as UI testing. While existing benchmarks evaluate FM-based UI navigation using the binary pass/fail metric, they have two major limitations: they cannot reflect the complex nature of mobile UI navigation where FMs may fail for various reasons (e.g., misunderstanding instructions and failed planning), and they lack industrial relevance due to oversimplified tasks that poorly represent real-world scenarios. To address the preceding limitations, we propose Sphinx, a comprehensive benchmark for multi-dimensional evaluation of FMs in industrial settings of UI navigation. Sphinx introduces a specialized toolkit that evaluates five essential FM capabilities, providing detailed insights into failure modes such as insufficient app knowledge or planning issues. Using both popular Google Play applications and WeChat's internal UI test cases, we evaluate 8 FMs with 20 different configurations. Our results show that existing FMs universally struggle with goal-based testing tasks, primarily due to insufficient UI-specific capabilities. We summarize seven lessons learned from benchmarking FMs with Sphinx, providing clear directions for improving FM-based mobile UI navigation.

cs.SE↗

arXiv2025

DeCon: Detecting Incorrect Assertions via Postconditions Generated by a Large Language Model

Recently, given the docstring for the target problem and the target function signature, large language models (LLMs) have been used not only to generate source code, but also to generate test cases, consisting of test inputs and assertions (e.g., in the form of checking an actual output against the expected output). However, as shown by our empirical study on assertions generated by four LLMs for the HumanEval benchmark, over 62% of the generated assertions are incorrect (i.e., failed on the ground-truth problem solution). To detect incorrect assertions (given the docstring and the target function signature along with a sample of example inputs and outputs), in this paper, we propose a new approach named DeCon to effectively detect incorrect assertions via LLM-generated postconditions for the target problem (a postcondition is a predicate that must always be true just after the execution of the ground-truth problem solution). Our approach requires a small set of I/O examples (i.e., a sample of example inputs and outputs) for the target problem (e.g., the I/O examples included in the docstring for a target problem in HumanEval). We use the given I/O examples to filter out those LLM-generated postconditions that are violated by at least one given I/O example. We then use the remaining postconditions to detect incorrect assertions as those assertions that violate at least one remaining postcondition. Experimental results show that DeCon can detect averagely more than 64% (63% and 65.5% detected by GPT-3.5 and GPT-4, respectively) incorrect assertions generated by four state-of-the-art LLMs, and DeCon can also improve the effectiveness of these LLMs in code generation by 4% in terms of Pass@1. In addition, although DeCon might filter out correct assertions, the fault-finding ability of the remaining correct assertions decreases only slightly.

cs.SE↗

arXiv2024

Preparing for Super-Reactivity: Early Fault-Detection in the Development of Exceedingly Complex Reactive Systems

We introduce the term Super-Reactive Systems to refer to reactive systems whose construction and behavior are complex, constantly changing and evolving, and heavily interwoven with other systems and the physical world. Finding hidden faults in such systems early in planning and development is critical for human safety, the environment, society and the economy. However, the complexity of the system and its interactions and the absence of adequate technical details pose a great obstacle. We propose an architecture for models and tools to overcome such barriers and enable simulation, systematic analysis, and fault detection and handling, early in the development of super-reactive systems. The approach is facilitated by the inference and abstraction capabilities and the power and knowledge afforded by large language models and associated AI tools. It is based on: (i) deferred, just-in-time interpretation of model elements that are stored in natural language form, and (ii) early capture of tacit interdependencies among seemingly orthogonal requirements.

cs.SE↗

arXiv2024

Distributed Speculative Inference (DSI): Speculation Parallelism for Provably Faster Lossless Language Model Inference

This paper introduces distributed speculative inference (DSI), a novel inference algorithm that is provably faster than speculative inference (SI) [leviathan2023, chen2023, miao2024, sun2025, timor2025] and standard autoregressive inference (non-SI). Like other SI algorithms, DSI operates on frozen language models (LMs), requiring no training or architectural modifications, and it preserves the target distribution. Prior studies on SI have demonstrated empirical speedups over non-SI--but rely on sufficiently fast and accurate drafters, which are often unavailable in practice. We identify a gap where SI can be slower than non-SI if drafters are too slow or inaccurate. We close this gap by proving that DSI is faster than both SI and non-SI--given any drafters. DSI is therefore not only faster than SI, but also unlocks the acceleration of LMs for which SI fails. DSI leverages speculation parallelism (SP), a novel type of task parallelism, to orchestrate target and drafter instances that overlap in time, establishing a new foundational tradeoff between computational resources and latency. Our simulations show that DSI is 1.29-1.92x faster than SI in single-node setups for various off-the-shelf LMs and tasks. We open-source all our code.

cs.DC↗

arXiv2024

Non-verbal information in spontaneous speech -- towards a new framework of analysis

Non-verbal signals in speech are encoded by prosody and carry information that ranges from conversation action to attitude and emotion. Despite its importance, the principles that govern prosodic structure are not yet adequately understood. This paper offers an analytical schema and a technological proof-of-concept for the categorization of prosodic signals and their association with meaning. The schema interprets surface-representations of multi-layered prosodic events. As a first step towards implementation, we present a classification process that disentangles prosodic phenomena of three orders. It relies on fine-tuning a pre-trained speech recognition model, enabling the simultaneous multi-class/multi-label detection. It generalizes over a large variety of spontaneous data, performing on a par with, or superior to, human annotation. In addition to a standardized formalization of prosody, disentangling prosodic patterns can direct a theory of communication and speech organization. A welcome by-product is an interpretation of prosody that will enhance speech- and language-related technologies.

cs.SD↗

arXiv2024

On Augmenting Scenario-Based Modeling with Generative AI

The manual modeling of complex systems is a daunting task; and although a plethora of methods exist that mitigate this issue, the problem remains very difficult. Recent advances in generative AI have allowed the creation of general-purpose chatbots, capable of assisting software engineers in various modeling tasks. However, these chatbots are often inaccurate, and an unstructured use thereof could result in erroneous system models. In this paper, we outline a method for the safer and more structured use of chatbots as part of the modeling process. To streamline this integration, we propose leveraging scenario-based modeling techniques, which are known to facilitate the automated analysis of models. We argue that through iterative invocations of the chatbot and the manual and automatic inspection of the resulting models, a more accurate system model can eventually be obtained. We describe favorable preliminary results, which highlight the potential of this approach.

cs.SE↗

arXiv2023

Toward Methodical Discovery and Handling of Hidden Assumptions in Complex Systems and Models

Methodologies for development of complex systems and models include external reviews by domain and technology experts. Among others, such reviews can uncover undocumented built-in assumptions that may be critical for correct and safe operation or constrain applicability. Since such assumptions may still escape human-centered processes like reviews, agile development, and risk analyses, here, we contribute toward making this process more methodical and automatable. We first present a blueprint for a taxonomy and formalization of the problem. We then show that a variety of digital artifacts of the system or model can be automatically checked against extensive reference knowledge. Since mimicking the breadth and depth of knowledge and skills of experts may appear unattainable, we illustrate the basic feasibility of automation with rudimentary experiments using OpenAI's ChatGPT. We believe that systematic handling of this aspect of system engineering can contribute significantly to the quality and safety of complex systems and models, and to the efficiency of development projects. We dedicate this work to Werner Damm, whose contributions to modeling and model-based development, in industry and academia, with a special focus on safety, helped establish a solid foundation to our discipline and to the work of many scientists and professionals, including, naturally, the approaches and techniques described here.

cs.SE↗

arXiv2023

Natural Averaging May Complement Known Biological Constraints in Bi-parental Reproduction's Advantages Over Mono-parental in Conserving Species Quantitative Traits

Commonly recognized evolutionarily relevant effects of sexual reproduction include increased diversity, accelerated adaptation, and constrained accumulation of deleterious mutations, along with a secondary effect of species genotype homogenization. Still, strong published arguments prioritize the contribution of biological mechanisms underlying bi-parental reproduction to maintaining species identity above their contribution to diversity. Here, we contribute to the latter position. In an initial mathematical analysis and simulation, we show that in an environment where copying is prone to error, quantitative polygenic traits that are shared within a parents' generation are transmitted to future generations under bi-parental reproduction with less deviation than under asexual reproduction. Furthermore, we abstract away many biological details, and show that this trait conservation is a general statistical effect, driven by the very nature of mixing of parental traits, separately from DNA repair and from the reproductive failures, barriers and disadvantages induced by biological mechanisms. Since survival of ecosystem interaction networks depends on the ability of individuals to replace the networked function of failing, dying or absent members of the same species, more faithful inheritance of common traits helps sustain species and ecosystems. This sustaining effect may have contributed to the very evolution of sexual reproduction.

q-bio.PE↗

arXiv2023

The Human-or-Machine Matter: Turing-Inspired Reflections on an Everyday Issue

In his seminal paper ``Computing Machinery and Intelligence'', Alan Turing introduced the ``imitation game'' as part of exploring the concept of machine intelligence. The Turing Test has since been the subject of much analysis, debate, refinement and extension. Here we sidestep the question of whether a particular machine can be labeled intelligent, or can be said to match human capabilities in a given context. Instead, we first draw attention to the seemingly simpler question a person may ask themselves in an everyday interaction: ``Am I interacting with a human or with a machine?''. We then shift the focus from seeking a method for eliciting the answer, and, rather, reflect upon the importance and significance of this Human-or-Machine question and the use one may make of a reliable answer thereto. Whereas Turing's original test is widely considered to be more of a thought experiment, the Human-or-Machine matter as discussed here has obvious practical relevance. While it is still unclear if and when machines will be able to mimic human behavior with high fidelity in everyday contexts, we argue that near-term exploration of the issues raised here can contribute to refinement of methods for developing computerized systems, and may also lead to new insights into fundamental characteristics of human behavior.

cs.AI↗

arXiv2023

Verifying Learning-Based Robotic Navigation Systems

Deep reinforcement learning (DRL) has become a dominant deep-learning paradigm for tasks where complex policies are learned within reactive systems. Unfortunately, these policies are known to be susceptible to bugs. Despite significant progress in DNN verification, there has been little work demonstrating the use of modern verification tools on real-world, DRL-controlled systems. In this case study, we attempt to begin bridging this gap, and focus on the important task of mapless robotic navigation -- a classic robotics problem, in which a robot, usually controlled by a DRL agent, needs to efficiently and safely navigate through an unknown arena towards a target. We demonstrate how modern verification engines can be used for effective model selection, i.e., selecting the best available policy for the robot in question from a pool of candidate policies. Specifically, we use verification to detect and rule out policies that may demonstrate suboptimal behavior, such as collisions and infinite loops. We also apply verification to identify models with overly conservative behavior, thus allowing users to choose superior policies, which might be better at finding shorter paths to a target. To validate our work, we conducted extensive experiments on an actual robot, and confirmed that the suboptimal policies detected by our method were indeed flawed. We also demonstrate the superiority of our verification-driven approach over state-of-the-art, gradient attacks. Our work is the first to establish the usefulness of DNN verification in identifying and filtering out suboptimal DRL policies in real-world robots, and we believe that the methods presented here are applicable to a wide range of systems that incorporate deep-learning-based agents.

cs.RO↗