# SteerEval: Inference-time Interventions Strengthen Multilingual Generalization in Neural Summarization Metrics

arXiv: https://arxiv.org/abs/2601.15809 · HTML: https://mlsanity.com/paper/2601.15809/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 29 of its 42 references that are in the map.
- First posted on arXiv 22 Jan 2026.

## Builds on (most cited first, 29 of 29)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [BLEURT: Learning Robust Metrics for Text Generation](https://mlsanity.com/paper/2004.04696/) (2020)
- [COMET: A Neural Framework for MT Evaluation](https://mlsanity.com/paper/2009.09025/) (2020)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [SummEval: Re-evaluating Summarization Evaluation](https://mlsanity.com/paper/2007.12626/) (2020)
- [The FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translation](https://mlsanity.com/paper/2106.03193/) (2021)
- [XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 Languages](https://mlsanity.com/paper/2106.13822/) (2021)
- [Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model](https://mlsanity.com/paper/2402.07827/) (2024)
- [Do Llamas Work in English? On the Latent Language of Multilingual Transformers](https://mlsanity.com/paper/2402.10588/) (2024)
- [Evaluation of Text Generation: A Survey](https://mlsanity.com/paper/2006.14799/) (2020)
- [LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks](https://mlsanity.com/paper/2406.18403/) (2024)
- [Finding Universal Grammatical Relations in Multilingual BERT](https://mlsanity.com/paper/2005.04511/) (2020)
- [Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers](https://mlsanity.com/paper/2411.08745/) (2024)
- [SEAHORSE: A Multilingual, Multifaceted Dataset for Summarization Evaluation](https://mlsanity.com/paper/2305.13194/) (2023)
- [Refusal Direction is Universal Across Safety-Aligned Languages](https://mlsanity.com/paper/2505.17306/) (2025)
- [Bridging the Language Gaps in Large Language Models with Inference-Time Cross-Lingual Intervention](https://mlsanity.com/paper/2410.12462/) (2024)
- [METAL: Towards Multilingual Meta-Evaluation](https://mlsanity.com/paper/2404.01667/) (2024)
- [MINERS: Multilingual Language Models as Semantic Retrievers](https://mlsanity.com/paper/2406.07424/) (2024)
- [The Translation Barrier Hypothesis: Multilingual Generation with Large Language Models Suffers from Implicit Translation Failure](https://mlsanity.com/paper/2506.22724/) (2025)
- [RepEval: Effective Text Evaluation with LLM Representation](https://mlsanity.com/paper/2404.19563/) (2024)
- [HeSum: a Novel Dataset for Abstractive Text Summarization in Hebrew](https://mlsanity.com/paper/2406.03897/) (2024)
- [ReCoVeR the Target Language: Language Steering without Sacrificing Task Performance](https://mlsanity.com/paper/2509.14814/) (2025)
- [Beyond N-Grams: Rethinking Evaluation Metrics and Strategies for Multilingual Abstractive Summarization](https://mlsanity.com/paper/2507.08342/) (2025)
- [Evaluating the Evaluator: Summarization Metrics and LLM-Judges beyond English](https://mlsanity.com/paper/2503.17039/) (2025)
- [References Matter: Investigating the Impact of Reference Set Variation on Summarization Evaluation](https://mlsanity.com/paper/2506.14335/) (2025)

## Abstract (arXiv)

An increasing body of work has leveraged multilingual language models for Natural Language Generation tasks such as summarization. A major empirical bottleneck in this area is the shortage of accurate and robust evaluation metrics for many languages, which hinders progress. Recent studies suggest that multilingual language models often use English as an internal pivot language, and that misalignment with this pivot can lead to degraded downstream performance. Motivated by the hypothesis that this mismatch could also apply to multilingual neural metrics, we ask whether steering their activations toward an English pivot can improve correlation with human judgments. We experiment with encoder- and decoder-based metrics and find that test-time intervention methods are effective across the board, increasing metric effectiveness for diverse languages.