arXiv subjects

Michelangelo Naim

Publications and source records attributed to Michelangelo Naim.

7 recordsLinked to original sources

arXiv2026

Pact: A Choreographic Language for Agentic Ecosystems

Recent advances in large language models have led to the rise of software systems (i.e. agents) that execute with increasing autonomy on behalf of users in open, multi-party settings, interacting with untrusted counterparts and managing private information. Choreographic programming offers correct-by-construction protocol-design for such settings, but assumes cooperative participants -- it has no notion of agent self-interest, that is, why an agent will follow a protocol. In this talk we introduce Pact, a choreographic language extended with operations to describe agent choices and preferences, drawing from the rich literature of game theory. Every Pact protocol maps to a formal game, allowing protocol designers to reason about game-theoretic properties of their protocols, such as solving for decision policies. We present Pact's design and a preliminary implementation -- a bounded-rational solver that computes decision policies over Pact protocols -- and findings from applying this language to multi-party coordination with self-interested agentic participants.

cs.PL↗

arXiv2025

Benchmarking World-Model Learning with Environment-Level Queries

World models are central to building AI agents capable of flexible reasoning and planning. Yet current evaluations (i) test only properties measurable from observed interactions, such as next-frame prediction or task return, and (ii) do not test whether a learned model supports diverse queries about the environment. In contrast, humans build $\textit{general-purpose}$ models that can answer many different questions about an environment$\unicode{x2014}$including questions that require understanding global structure and counterfactual consequences. We propose $\textit{WorldTest}$: a protocol for evaluating whether agents learn models that support multiple $\textit{environment-level queries}\unicode{x2014}$questions whose answers depend on properties of the full environment, not just observed trajectories. Individually, these queries can target properties (e.g., reachability or the effects of interventions) that no single rollout distribution determines. Collectively, they assess model generality across query types. We instantiate WorldTest as $\textit{AutumnBench}$, a benchmark of 43 interactive grid-world environments and 129 tasks across three query families for both humans and learning agents. Experiments with 517 human participants and five frontier models show that humans substantially outperform these models, a gap we attribute to differences in exploration and belief updating. AutumnBench provides a framework for evaluating world-model learning in grid-world environments with environment-level queries, and WorldTest provides a template for extending such evaluations to richer domains.

cs.AI↗

arXiv2024

Combining Induction and Transduction for Abstract Reasoning

When learning an input-output mapping from very few examples, is it better to first infer a latent function that explains the examples, or is it better to directly predict new test outputs, e.g. using a neural network? We study this question on ARC by training neural models for induction (inferring latent functions) and transduction (directly predicting the test output for a given test input). We train on synthetically generated variations of Python programs that solve ARC training tasks. We find inductive and transductive models solve different kinds of test problems, despite having the same training problems and sharing the same neural architecture: Inductive program synthesis excels at precise computations, and at composing multiple concepts, while transduction succeeds on fuzzier perceptual concepts. Ensembling them approaches human-level performance on ARC.

cs.LG↗

arXiv2024

Baba Is AI: Break the Rules to Beat the Benchmark

Humans solve problems by following existing rules and procedures, and also by leaps of creativity to redefine those rules and objectives. To probe these abilities, we developed a new benchmark based on the game Baba Is You where an agent manipulates both objects in the environment and rules, represented by movable tiles with words written on them, to reach a specified goal and win the game. We test three state-of-the-art multi-modal large language models (OpenAI GPT-4o, Google Gemini-1.5-Pro and Gemini-1.5-Flash) and find that they fail dramatically when generalization requires that the rules of the game must be manipulated and combined.

cs.CL↗

arXiv2023

Lyfe Agents: Generative agents for low-cost real-time social interactions

Highly autonomous generative agents powered by large language models promise to simulate intricate social behaviors in virtual societies. However, achieving real-time interactions with humans at a low computational cost remains challenging. Here, we introduce Lyfe Agents. They combine low-cost with real-time responsiveness, all while remaining intelligent and goal-oriented. Key innovations include: (1) an option-action framework, reducing the cost of high-level decisions; (2) asynchronous self-monitoring for better self-consistency; and (3) a Summarize-and-Forget memory mechanism, prioritizing critical memory items at a low cost. We evaluate Lyfe Agents' self-motivation and sociability across several multi-agent scenarios in our custom LyfeGame 3D virtual environment platform. When equipped with our brain-inspired techniques, Lyfe Agents can exhibit human-like self-motivated social reasoning. For example, the agents can solve a crime (a murder mystery) through autonomous collaboration and information exchange. Meanwhile, our techniques enabled Lyfe Agents to operate at a computational cost 10-100 times lower than existing alternatives. Our findings underscore the transformative potential of autonomous generative agents to enrich human social experiences in virtual worlds.

cs.HC↗

arXiv2020

Fundamental Law of Memory Recall

Free recall of random lists of words is a standard paradigm used to probe human memory. We proposed an associative search process that can be reduced to a deterministic walk on random graphs defined by the structure of memory representations. This model makes a parameter-free prediction for the average number of memory items recalled ($RC$) out of $M$ items in memory: $R = \sqrt{3πM/2}$. This prediction was verified in a large-scale crowd-sourced free recall and recognition experiment. We uncovered a novel law of memory recall, indicating that recall operates according to a stereotyped search process common to all people.

q-bio.NC↗

arXiv2018

Reducing a cortical network to a Potts model yields storage capacity estimates

An autoassociative network of Potts units, coupled via tensor connections, has been proposed and analysed as an effective model of an extensive cortical network with distinct short- and long-range synaptic connections, but it has not been clarified in what sense it can be regarded as an effective model. We draw here the correspondence between the two, which indicates the need to introduce a local feedback term in the reduced model, i.e., in the Potts network. An effective model allows the study of phase transitions. As an example, we study the storage capacity of the Potts network with this additional term, the local feedback $w$, which contributes to drive the activity of the network towards one of the stored patterns. The storage capacity calculation, performed using replica tools, is limited to fully connected networks, for which a Hamiltonian can be defined. To extend the results to the case of intermediate partial connectivity, we also derive the self-consistent signal-to-noise analysis for the Potts network; and finally we discuss implications for semantic memory in humans.

q-bio.NC↗