arXiv subjects

Ignacio Sastre

Publications and source records attributed to Ignacio Sastre.

7 recordsLinked to original sources

arXiv2026

Derivation Prompting: A Logic-Based Method for Improving Retrieval-Augmented Generation

The application of Large Language Models to Question Answering has shown great promise, but important challenges such as hallucinations and erroneous reasoning arise when using these models, particularly in knowledge-intensive, domain-specific tasks. To address these issues, we introduce Derivation Prompting, a novel prompting technique for the generation step of the Retrieval-Augmented Generation framework. Inspired by logic derivations, this method involves deriving conclusions from initial hypotheses through the systematic application of predefined rules. It constructs a derivation tree that is interpretable and adds control over the generation process. We applied this method in a specific case study, significantly reducing unacceptable answers compared to traditional RAG and long-context window methods.

cs.CL↗

arXiv2026

RETUYT-INCO at BEA 2026 Shared Task 2: Meta-prompting in Rubric-based Scoring for German

In this paper, we present the RETUYT-INCO participation at the BEA 2026 shared task "Rubric-based Short Answer Scoring for German". Our team participated in track 1 (Unseen answers three-way), track 3 (Unseen answers two-way) and track 4 (Unseen questions two-way). Since these tracks required scoring short student answers using specific rubrics, we looked for ways to handle the changing nature of the task. We created a method called Meta-prompting. In this approach, an LLM creates a custom prompt based on examples from the Train set. This prompt is then used to grade new student answers. Along with this method, we also describe other approaches we used, such as classic machine learning, fine-tuning open-source LLMs, and different prompting techniques. According to the official results, our team placed 6th out of 8 participants in Track 1 with a QWK of 0.729. In Track 3, we secured 4th place out of 9 with a QWK of 0.674, and we also placed 4th out of 8 in Track 4 with a QWK of 0.49.

cs.CL↗

arXiv2026

Concept Tokens: Learning Behavioral Embeddings Through Concept Definitions

We propose Concept Tokens, a lightweight method that adds a new special token to a pretrained LLM and learns only its embedding from multiple natural language definitions of a target concept, where occurrences of the concept are replaced by the new token. The LLM is kept frozen and the embedding is optimized with the standard language-modeling objective. We evaluate Concept Tokens in three settings. First, we study hallucinations in closed-book question answering on HotpotQA and find a directional effect: negating the hallucination token reduces hallucinated answers mainly by increasing abstentions, whereas asserting it increases hallucinations and lowers precision. Second, we induce recasting, a pedagogical feedback strategy for second language teaching, and observe the same directional effect. Moreover, compared to providing the full definitional corpus in-context, concept tokens better preserve compliance with other instructions (e.g., asking follow-up questions). Finally, we include a qualitative study with the Eiffel Tower and a fictional "Austral Tower" to illustrate what information the learned embeddings capture and where their limitations emerge. Overall, Concept Tokens provide a compact control signal learned from definitions that can steer behavior in frozen LLMs.

cs.CL↗

arXiv2025

On the Use of Large Language Models for Qualitative Synthesis

Large language models (LLMs) show promise for supporting systematic reviews (SR), even complex tasks such as qualitative synthesis (QS). However, applying them to a stage that is unevenly reported and variably conducted carries important risks: misuse can amplify existing weaknesses and erode confidence in the SR findings. To examine the challenges of using LLMs for QS, we conducted a collaborative autoethnography involving two trials. We evaluated each trial for methodological rigor and practical usefulness, and interpreted the results through a technical lens informed by how LLMs are built and their current limitations.

cs.SE↗

arXiv2025

Memory Tokens: Large Language Models Can Generate Reversible Sentence Embeddings

In this work, we observe an interesting phenomenon: it is possible to generate reversible sentence embeddings that allow an LLM to reconstruct the original text exactly, without modifying the model's weights. This is achieved by introducing a special memory token, whose embedding is optimized through training on a fixed sequence. When prompted with this embedding, the model reconstructs the fixed sequence exactly. We evaluate this phenomenon across English and Spanish datasets, sequences of up to approximately 240 tokens, and model scales ranging from 100M to 8B parameters. Notably, Llama 3.1 8B successfully reconstructs all tested sequences. Our findings highlight an interesting capability of LLMs and suggest potential applications in memory-based retrieval, compression, and controlled text generation.

cs.CL↗

arXiv2025

RETUYT-INCO at BEA 2025 Shared Task: How Far Can Lightweight Models Go in AI-powered Tutor Evaluation?

In this paper, we present the RETUYT-INCO participation at the BEA 2025 shared task. Our participation was characterized by the decision of using relatively small models, with fewer than 1B parameters. This self-imposed restriction tries to represent the conditions in which many research labs or institutions are in the Global South, where computational power is not easily accessible due to its prohibitive cost. Even under this restrictive self-imposed setting, our models managed to stay competitive with the rest of teams that participated in the shared task. According to the $exact\ F_1$ scores published by the organizers, the performance gaps between our models and the winners were as follows: $6.46$ in Track 1; $10.24$ in Track 2; $7.85$ in Track 3; $9.56$ in Track 4; and $13.13$ in Track 5. Considering that the minimum difference with a winner team is $6.46$ points -- and the maximum difference is $13.13$ -- according to the $exact\ F_1$ score, we find that models with a size smaller than 1B parameters are competitive for these tasks, all of which can be run on computers with a low-budget GPU or even without a GPU.

cs.CL↗

arXiv2025

A Platform for Generating Educational Activities to Teach English as a Second Language

We present a platform for the generation of educational activities oriented to teaching English as a foreign language. The different activities -- games and language practice exercises -- are strongly based on Natural Language Processing techniques. The platform offers the possibility of playing out-of-the-box games, generated from resources created semi-automatically and then manually curated. It can also generate games or exercises of greater complexity from texts entered by teachers, providing a stage of review and edition of the generated content before use. As a way of expanding the variety of activities in the platform, we are currently experimenting with image and text generation. In order to integrate them and improve the performance of other neural tools already integrated, we are working on migrating the platform to a more powerful server. In this paper we describe the development of our platform and its deployment for end users, discussing the challenges faced and how we overcame them, and also detail our future work plans.

cs.CL↗