Search Results for author: Luciano del Corro

Found 17 papers, 3 papers with code

Are Optimal Algorithms Still Optimal? Rethinking Sorting in LLM-Based Pairwise Ranking with Batching and Caching

no code implementations30 May 2025 Juan Wisznia, Cecilia Bolaños, Juan Tollo, Giovanni Marraffini, Agustín Gianolini, Noe Hsueh, Luciano del Corro

We introduce a novel framework for analyzing sorting algorithms in pairwise ranking prompting (PRP), re-centering the cost model around LLM inferences rather than traditional pairwise comparisons.

The Greatest Good Benchmark: Measuring LLMs' Alignment with Utilitarian Moral Dilemmas

no code implementations25 Mar 2025 Giovanni Franco Gabriel Marraffini, Andrés Cotton, Noe Fabian Hsueh, Axel Fridman, Juan Wisznia, Luciano del Corro

The question of how to make decisions that maximise the well-being of all persons is very relevant to design language models that are beneficial to humanity and free from harm.

sPhinX: Sample Efficient Multilingual Instruction Fine-Tuning Through N-shot Guided Prompting

no code implementations13 Jul 2024 Sanchit Ahuja, Kumar Tanmay, Hardik Hansrajbhai Chauhan, Barun Patra, Kriti Aggarwal, Luciano del Corro, Arindam Mitra, Tejas Indulal Dhamecha, Ahmed Awadallah, Monojit Choudhary, Vishrav Chaudhary, Sunayana Sitaram

In order to address this, we introduce a novel recipe for creating a multilingual synthetic instruction tuning dataset, sPhinX, which is created by selectively translating instruction response pairs from English into 50 languages.

Machine Translation Question Answering +1

AgentInstruct: Toward Generative Teaching with Agentic Flows

no code implementations3 Jul 2024 Arindam Mitra, Luciano del Corro, Guoqing Zheng, Shweti Mahajan, Dany Rouhana, Andres Codas, Yadong Lu, Wei-Ge Chen, Olga Vrousgos, Corby Rosset, Fillipe Silva, Hamed Khanpour, Yash Lara, Ahmed Awadallah

We focus on using synthetic data for post-training, specifically creating data by powerful models to teach a new skill or behavior to another model, we refer to this setting as Generative Teaching.

GSM8K MMLU +1

Automatic Pair Construction for Contrastive Post-training

1 code implementation3 Oct 2023 Canwen Xu, Corby Rosset, Ethan C. Chau, Luciano del Corro, Shweti Mahajan, Julian McAuley, Jennifer Neville, Ahmed Hassan Awadallah, Nikhil Rao

Remarkably, our automatic contrastive post-training further improves the performance of Orca, already a state-of-the-art instruction learning model tuned with GPT-4 outputs, to outperform ChatGPT.

SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inference

no code implementations5 Jul 2023 Luciano del Corro, Allie Del Giorno, Sahaj Agarwal, Bin Yu, Ahmed Awadallah, Subhabrata Mukherjee

While existing token-level early exit methods show promising results for online inference, they cannot be readily applied for batch inferencing and Key-Value caching.

Text Generation

Facts That Matter

1 code implementation EMNLP 2018 Marco Ponza, Luciano del Corro, Gerhard Weikum

This work introduces fact salience: The task of generating a machine-readable representation of the most prominent information in a text document as a set of facts.

Clustering Diversity +5

MinIE: Minimizing Facts in Open Information Extraction

1 code implementation EMNLP 2017 Kiril Gashteovski, Rainer Gemulla, Luciano del Corro

The goal of Open Information Extraction (OIE) is to extract surface relations and their arguments from natural-language text in an unsupervised, domain-independent manner.

Open Information Extraction Question Answering +1

Cannot find the paper you are looking for? You can Submit a new open access paper.