benchmark

multilingual mathematical benchmarks

benchmarkactivemultilingual-mathematical-benchmarks-9fc33934·1 events·first seen 25d ago

Aliases: multilingual mathematical benchmarks

Co-occurring entities

large language models LANG progressive decay schedule language-adaptive switch Reinforcement Learning

More like this (12)

MATH benchmark SWE-Bench Multilingual Mathlib mathematical reasoning Multilingual E5 CORE benchmark Mathstral 7B LCB/hard benchmark NuminaMath OpenSCAD Architectural 3D LLM Benchmark temporally grounded QA benchmark MLE-bench

Recent events (1)

6arXiv · cs.CL·25d ago·source ↗

LANG: Reinforcement Learning Framework for Multilingual Reasoning with Language-Adaptive Hint Guidance

LANG is a new RL-based framework for improving multilingual reasoning in LLMs that addresses the trade-off between input-language consistency and reasoning quality. It uses language-conditioned hints with a progressive decay schedule and a language-adaptive switch to tailor learning to per-language difficulty. Empirical results on multilingual mathematical benchmarks show improved reasoning without language drift toward English, and the approach generalizes beyond mathematics.

Evaluation and Benchmarking Alignment and RLHF large language models LANG multilingual mathematical benchmarks +3 more