arXiv subjects

Sergio Yovine

Publications and source records attributed to Sergio Yovine.

5 recordsLinked to original sources

arXiv2026

Minimality of Random Moore Automata under Prefix-Dependent Congruences

We study prefix-dependent congruences for random deterministic transition systems with state outputs. In this setting, the admissible continuations used to compare two states may depend on the observed prefix, and two states are identified only if no common admissible continuation distinguishes their future outputs. The framework includes probabilistic deterministic finite automata as a motivating special case. We analyze the random transition model in which all transition values are independent and uniform. Each state is also assigned an independent label that specifies both its output and its set of admissible symbols. If two independent labels agree with probability strictly less than one, and every label has at least three admissible symbols, then the induced congruence is trivial with high probability. The proof combines a pruning process on pairs, a collision-free exploration controlling its early evolution, and a first-moment argument showing that the remaining pairs cannot organize into nontrivial equivalence classes.

cs.FL↗

arXiv2026

TDAD: Test-Driven Agentic Development - Reducing Code Regressions in AI Coding Agents via Graph-Based Impact Analysis

AI coding agents can resolve real-world software issues, yet they frequently introduce regressions -- breaking tests that previously passed. Current benchmarks focus almost exclusively on resolution rate, leaving regression behavior under-studied. This paper presents TDAD (Test-Driven Agentic Development), an open-source tool that performs pre-change impact analysis for AI coding agents. TDAD builds a dependency map between source code and tests so that before committing a patch, the agent knows which tests to verify and can self-correct. The map is delivered as a lightweight agent skill -- a static text file the agent queries at runtime. Evaluated on SWE-bench Verified with two open-weight models running on consumer hardware (Qwen3-Coder 30B, 100 instances; Qwen3.5-35B-A3B, 25 instances), TDAD reduced regressions by 70% (6.08% to 1.82%) compared to a vanilla baseline. In contrast, adding TDD procedural instructions without targeted test context increased regressions to 9.94% -- worse than no intervention at all. When deployed as an agent skill with a different model and framework, TDAD improved issue-resolution rate from 24% to 32%, confirming that surfacing contextual information outperforms prescribing procedural workflows. All code, data, and logs are publicly available at https://github.com/pepealonso95/TDAD.

cs.SE↗

arXiv2024

Congruence-based Learning of Probabilistic Deterministic Finite Automata

This work studies the question of learning probabilistic deterministic automata from language models. For this purpose, it focuses on analyzing the relations defined on algebraic structures over strings by equivalences and similarities on probability distributions. We introduce a congruence that extends the classical Myhill-Nerode congruence for formal languages. This new congruence is the basis for defining regularity over language models. We present an active learning algorithm that computes the quotient with respect to this congruence whenever the language model is regular. The paper also defines the notion of recognizability for language models and shows that it coincides with regularity for congruences. For relations which are not congruences, it shows that this is not the case. Finally, it discusses the impact of this result on learning in the context of language models.

cs.FL↗

arXiv2024

Analyzing constrained LLM through PDFA-learning

We define a congruence that copes with null next-symbol probabilities that arise when the output of a language model is constrained by some means during text generation. We develop an algorithm for efficiently learning the quotient with respect to this congruence and evaluate it on case studies for analyzing statistical properties of LLM.

cs.FL↗

arXiv2022

Towards Efficient Active Learning of PDFA

We propose a new active learning algorithm for PDFA based on three main aspects: a congruence over states which takes into account next-symbol probability distributions, a quantization that copes with differences in distributions, and an efficient tree-based data structure. Experiments showed significant performance gains with respect to reference implementations.

cs.FL↗