LetTokenBurn
Sign in
Auto71%Harbor: pendingMultimodalKnowledge & Reasoning

DailyClue

DailyClue Dataset Seek-and-Solve: Benchmarking MLLMs for Visual Clue-Driven Reasoning in Daily Scenarios Dataset Summary DailyClue is a benchmark for evaluating visual clue-driven reasoning in Multimodal Large Language Models (MLLMs).

PaperDataset

How to run

ltb · dailyclue
$ npm i -g lettokenburn-cli
$ ltb bench info dailyclue

External harness. No Harbor adapter has been generated for this benchmark yet. `ltb bench info` prints the known links.

Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs

Scores

0 sourced entries

No sourced scores recorded yet.

Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.

Reproduced by the community

All runs
Loading community runs…

About

DailyClue Dataset

Seek-and-Solve: Benchmarking MLLMs for Visual Clue-Driven Reasoning in Daily Scenarios

Dataset Summary

DailyClue is a benchmark for evaluating visual clue-driven reasoning in Multimodal Large Language Models (MLLMs). Unlike benchmarks that test pre-existing knowledge, DailyClue requires models to actively identify decisive visual clues from images before producing answers. The dataset spans 4 major domains and 16 distinct subtasks, with 666 total…