Recent Accepted Papers (2026): PROPOR, AISTATS, and ICLR Workshop
Table of Contents
I am very happy to share three recent accepted publications from 2026:
1. MATH-PT: A Math Reasoning Benchmark for European and Brazilian Portuguese #
Authors: Tiago Teixeira, Ana Carolina Erthal, Juan Belieni, Beatriz Canaverde, Miguel Faria, Diego Mesquita, Eliezer de Souza da Silva, André Martins
Venue: 17th Conference on Computational Processing of Portuguese (PROPOR 2026)
Summary: We introduce MATH-PT, a benchmark of 1,729 native Portuguese math problems (European and Brazilian Portuguese) designed to evaluate mathematical reasoning in modern LLMs.
2. On the Identifiability of Tensor Ranks via Prior Predictive Matching #
Authors: Eliezer de Souza da Silva, Arto Klami, Diego Mesquita, Iñigo Urteaga
Venue: Artificial Intelligence and Statistics (AISTATS 2026)
Summary: This paper presents a prior-predictive-matching framework for studying identifiability of tensor ranks, showing which tensor models admit identifiable ranks and deriving closed-form estimators for identifiable cases.
3. Orthogonal Gradient Projection for Continual LLM Unlearning #
Authors: Juan Belieni, Ana Carolina Erthal, Eliezer de Souza da Silva, Diego Mesquita
Venue: ICLR 2026 Workshop on AI with Recursive Self-Improvement
Summary: We present a method based on orthogonal gradient projection for continual LLM unlearning in recursive self-improvement scenarios.
More updates soon!