Skip to content

Commit

Permalink
fix: update convention to align pre-commit
Browse files Browse the repository at this point in the history
  • Loading branch information
honghanhh committed Oct 28, 2024
1 parent 71fd193 commit 18ccd90
Showing 1 changed file with 17 additions and 6 deletions.
23 changes: 17 additions & 6 deletions lib/shroom/README.md
Original file line number Diff line number Diff line change
@@ -1,29 +1,40 @@
# SemEval-2025 Task-3 — Mu-SHROOM, the Multilingual Shared-task on Hallucinations and Related Observable Overgeneration Mistakes

[Description](#description) | [Data](#data) | [Models](#models) | [Results](#results) |
[Leaderboard](#leaderboard) |[Contributors](#contributors)

[Description](#description) | [Data](#data) | [Models](#models) | [Results](#results) | [Leaderboard](#leaderboard) |[Contributors](#contributors)

In this repo, we provide our solution to solve [Mu-SHROOM, the Multilingual Shared-task on Hallucinations and Related Observable Overgeneration Mistakes](https://helsinki-nlp.github.io/shroom/).
In this repo, we provide our solution to solve
[Mu-SHROOM, the Multilingual Shared-task on Hallucinations and Related Observable Overgeneration Mistakes](https://helsinki-nlp.github.io/shroom/).

## Description

ARKHN aims to detect hallucination spans in the outputs of instruction-tuned LLMs in a multilingual context in Mu-SHROOM, which stands for “Multilingual Shared-task on Hallucinations and Related Observable Overgeneration Mistakes”.
ARKHN aims to detect hallucination spans in the outputs of instruction-tuned LLMs in a multilingual
context in Mu-SHROOM, which stands for “Multilingual Shared-task on Hallucinations and Related
Observable Overgeneration Mistakes”.

Evaluation metrics:
- intersection-over-union of characters marked as hallucinations in the gold reference vs. predicted as such
- how well the probability assigned by the participants’ system that a character is part of a hallucination correlates with the empirical probabilities observed in our annotators.

- intersection-over-union of characters marked as hallucinations in the gold reference vs. predicted
as such
- how well the probability assigned by the participants’ system that a character is part of a
hallucination correlates with the empirical probabilities observed in our annotators.

## Data

(updating)

## Models

(updating)

## Results

(updating)

## Leaderboard

(updating)

## Contributors

(updating)

0 comments on commit 18ccd90

Please sign in to comment.