Update README.md: enhance experiments section with detailed scoring and tool descriptions

README.md

@@ -48,18 +48,22 @@ uv run gradio app.py
 ```
 
 # experiments
+The score is max score after a few runs
 | Features | Model | Score (out of 20) | Notes |
 | --- | --- | --- | --- |
-| Directly call LLM with questions | gemini-flash 2.0 | 1 | direct call |
-| + Structured Output: Response Format | gemini-flash 2.0 | 3 |  |
-| + system prompt from GAIA, slightly modified | gemini-flash 2.0 | 3 |  |
-| + Agentic LLM with Smolagent & Add “Understand video” tool using gemini-flash. | gemini-flash 2.0 | 1-4 | Results varied due to `Codeagent` output formatting issues. |
-| + WebSearch tool | gemini-flash 2.0 | 7 | Greatly improved, still has formatting issues |
-| + Change model | gemini-flash 2.5 | 7 |  |
-| + VisitWebpageTool | gemini-flash 2.5 | 10-12 | cannot see image, mp3, excel file. |
+| Prompted LLM | gemini-flash 2.0 | 1 | Format could be wrong |
+| + Structured Output | gemini-flash 2.0 | 3 |  |
+| + System prompt from GAIA, slightly modified | gemini-flash 2.0 | 3 |  |
+| + Agentic LLM with Smolagent & Add “Understand video” tool using gemini-flash. | gemini-flash 2.0 | 4 | Results varied due to `Codeagent` output formatting issues. |
+| + WebSearchTool | gemini-flash 2.0 | 7 | Greatly improved, still has formatting issues |
+| + Change from flash 2.0 to flash 2.5 | gemini-flash 2.5 | 7 |  |
+| + VisitWebpageTool | gemini-flash 2.5 | 10-12 | Cannot see image, mp3, excel file. |
 | + DownloadFile & ReadExcelFileBytes | gemini-flash 2.5 | 13 |  |
-|  |  |  |  |
-
+| + TranscribeAudioBytes | gemini-flash 2.5 | 14 |  |
+| + CodeExecutionTool | gemini-flash 2.5 | 14 |  |
+| + WikipediaSearchTool (customised using wikipedia + markdownify) | gemini-flash 2.5 | 15 | Tested both WikipediaRetriever and WikipediaLoader from LangChain, but their performance was limited — those omit table data, which can include key information. |
+| + Increase thinking budget on flash2.5 for performance & Edited system prompt | gemini-flash 2.5, temperature = 0 for consistency. | 15-17 | Stability is the key challenge. Can achieve 17 with GAIA score function, but this assignment is using exact match, so formatter needs improvement. |
+| + chess best move tool & Change Image understanding model to gemini-pro 2.5 | gemini-flash 2.5 & gemini-pro 2.5 | 18 | conversion of digital chessboard images to FEN strings is unstable. |
 
 # Developer
 ## Install pre-commit