CLAINov 2, 2025

The Riddle of Reflection: Evaluating Reasoning and Self-Awareness in Multilingual LLMs using Indian Riddles

arXiv:2511.00960v21 citationsh-index: 1
Originality Synthesis-oriented
AI Analysis

This work addresses the problem of culturally grounded reasoning and self-awareness in multilingual LLMs for researchers and developers, though it is incremental as it applies existing evaluation methods to new languages and data.

This paper evaluated the reasoning and self-assessment abilities of five large language models across seven Indian languages using a multilingual riddle dataset, finding that top-performing models like Gemini 2.5 Pro were overconfident with a 4.34% True Negative Rate, while lower-performing models like LLaMA 4 Scout showed greater self-awareness with a 42.09% True Negative Rate.

The extent to which large language models (LLMs) can perform culturally grounded reasoning across non-English languages remains underexplored. This paper examines the reasoning and self-assessment abilities of LLMs across seven major Indian languages-Bengali, Gujarati, Hindi, Kannada, Malayalam, Tamil, and Telugu. We introduce a multilingual riddle dataset combining traditional riddles with context-reconstructed variants and evaluate five LLMs-Gemini 2.5 Pro, Gemini 2.5 Flash, Mistral-Saba, LLaMA 4 Scout, and LLaMA 4 Maverick-under seven prompting strategies. In the first stage, we assess riddle-solving performance and find that while Gemini 2.5 Pro performs best overall, few-shot methods yield only marginal gains, and accuracy varies notably across languages. In the second stage, we conduct a self-evaluation experiment to measure reasoning consistency. The results reveal a key finding: a model's initial accuracy is inversely correlated with its ability to identify its own mistakes. Top-performing models such as Gemini 2.5 Pro are overconfident (4.34% True Negative Rate), whereas lower-performing models like LLaMA 4 Scout are substantially more self-aware (42.09% True Negative Rate). These results point to clear gaps in multilingual reasoning and highlight the need for models that not only reason effectively but also recognize their own limitations.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes