Cover Image

A gentle push funziona benissimo: making instructed models in Italian via contrastive activation steering

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Scalena, Daniel, Fersini, Elisabetta, Nissim, Malvina
Format:	Preprint
Published:	2024
Subjects:	Computation and Language Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Steering Large Language Models for Machine Translation Personalization
by: Scalena, Daniel, et al.
Published: (2025)

Multi-property Steering of Large Language Models with Dynamic Activation Composition
by: Scalena, Daniel, et al.
Published: (2024)

IT5: Text-to-text Pretraining for Italian Language Understanding and Generation
by: Sarti, Gabriele, et al.
Published: (2022)

Quantifying the Plausibility of Context Reliance in Neural Machine Translation
by: Sarti, Gabriele, et al.
Published: (2023)

Puzzled By ChatGPT? No more! A Jigsaw Puzzle to Promote AI Literacy and Awareness
by: Padovani, Francesca, et al.
Published: (2026)

Toward universal steering and monitoring of AI models
by: Beaglehole, Daniel, et al.
Published: (2025)

mCoT: Multilingual Instruction Tuning for Reasoning Consistency in Language Models
by: Lai, Huiyuan, et al.
Published: (2024)

Non Verbis, Sed Rebus: Large Language Models are Weak Solvers of Italian Rebuses
by: Sarti, Gabriele, et al.
Published: (2024)

TACLer: Tailored Curriculum Reinforcement Learning for Efficient Reasoning
by: Lai, Huiyuan, et al.
Published: (2026)

Efficient semantic uncertainty quantification in language models via diversity-steered sampling
by: Park, Ji Won, et al.
Published: (2025)

Choosy Babies Need One Coach: Inducing Mode-Seeking Behavior in BabyLlama with Reverse KL Divergence
by: Shi, Shaozhen, et al.
Published: (2024)

Multidimensional Consistency Improves Reasoning in Language Models
by: Lai, Huiyuan, et al.
Published: (2025)

When Harry Meets Superman: The Role of The Interlocutor in Persona-Based Dialogue Generation
by: Occhipinti, Daniela, et al.
Published: (2025)

Language steering in latent space to mitigate unintended code-switching
by: Goncharov, Andrey, et al.
Published: (2025)

Practising responsibility: Ethics in NLP as a hands-on course
by: Nissim, Malvina, et al.
Published: (2025)

Can Model Uncertainty Function as a Proxy for Multiple-Choice Question Item Difficulty?
by: Zotos, Leonidas, et al.
Published: (2024)

Are You Doubtful? Oh, It Might Be Difficult Then! Exploring the Use of Model Uncertainty for Question Difficulty Estimation
by: Zotos, Leonidas, et al.
Published: (2024)

The Role of the Availability Heuristic in Multiple-Choice Answering Behaviour
by: Zotos, Leonidas, et al.
Published: (2026)

JMI at SemEval 2024 Task 3: Two-step approach for multimodal ECAC using in-context learning with GPT and instruction-tuned Llama models
by: Arefa, et al.
Published: (2024)

Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models
by: Romero, Miguel, et al.
Published: (2025)

Can sparse autoencoders be used to decompose and interpret steering vectors?
by: Mayne, Harry, et al.
Published: (2024)

BLEUBERI: BLEU is a surprisingly effective reward for instruction following
by: Chang, Yapei, et al.
Published: (2025)

Isolating authorship from content with semantic embeddings and contrastive learning
by: Huertas-Tato, Javier, et al.
Published: (2024)

AI-University: An LLM-based platform for instructional alignment to scientific classrooms
by: Shojaei, Mostafa Faghih, et al.
Published: (2025)

Post-training makes large language models less human-like
by: Binz, Marcel, et al.
Published: (2026)

Unsupervised Word-level Quality Estimation for Machine Translation Through the Lens of Annotators (Dis)agreement
by: Sarti, Gabriele, et al.
Published: (2025)

Enabling robots to follow abstract instructions and complete complex dynamic tasks
by: Mon-Williams, Ruaridh, et al.
Published: (2024)

The ProLiFIC dataset: Leveraging LLMs to Unveil the Italian Lawmaking Process
by: Contestabile, Matilde, et al.
Published: (2025)

Round and Round We Go! What makes Rotary Positional Encodings useful?
by: Barbero, Federico, et al.
Published: (2024)

What makes a word hard to learn? Modeling L1 influence on English vocabulary difficulty
by: Martins, Jonas Mayer, et al.
Published: (2026)

Non-instructional Fine-tuning: Enabling Instruction-Following Capabilities in Pre-trained Language Models without Instruction-Following Data
by: Xie, Juncheng, et al.
Published: (2024)

Overcoming linguistic barriers in code assistants: creating a QLoRA adapter to improve support for Russian-language code writing instructions
by: Pronin, C. B., et al.
Published: (2024)

Evaluating Multimodal Language Models as Visual Assistants for Visually Impaired Users
by: Karamolegkou, Antonia, et al.
Published: (2025)

tinyCLAP: Distilling Constrastive Language-Audio Pretrained Models
by: Paissan, Francesco, et al.
Published: (2023)

Improving embedding with contrastive fine-tuning on small datasets with expert-augmented scores
by: Lu, Jun, et al.
Published: (2024)

Fine-tuning with HED-IT: The impact of human post-editing for dialogical language models
by: Occhipinti, Daniela, et al.
Published: (2024)

Modeling citation worthiness by using attention-based bidirectional long short-term memory networks and interpretable models
by: Zeng, Tong, et al.
Published: (2024)

Rewarding Graph Reasoning Process makes LLMs more Generalized Reasoners
by: Peng, Miao, et al.
Published: (2025)

SemCAFE: When Named Entities make the Difference Assessing Web Source Reliability through Entity-level Analytics
by: Shahi, Gautam Kishore, et al.
Published: (2025)

Small sample-based adaptive text classification through iterative and contrastive description refinement
by: Rajeev, Amrit, et al.
Published: (2025)