Add gpt-oss-120b leaderboard submission

[radiangle]

Summary

Adding evaluation results for gpt-oss-120b model to the leaderboard.

Results (4 trials per domain)

Domain	Pass@1	Pass@2	Pass@3	Pass@4
Airline	46.0%	35.3%	29.5%	26.0%
Retail	57.5%	45.8%	39.9%	36.0%
Telecom	58.1%	44.0%	36.6%	31.6%

Submission Details

• Model: gpt-oss-120b
• Model Organization: OpenAI
• Submitting Organization: Nori
• User Simulator: openai/gpt-oss-120b
• Trials per domain: 4
• Framework version: tau2-bench 0.2.1-dev

Agent Implementation

LangChain agent using LangGraph's create_react_agent. Implementation available at: https://github.com/radiangle/tau2-bench/blob/main/src/tau2/agent/langchain_agent.py

Verification

• No modifications to prompts
• No tasks omitted
• Standard tau2-bench evaluation protocol

[victorb-sierra]

Thank you for your submission.
cc: @benshi34

[benshi34]

Hi: your trajectory file names are formatted incorrectly, would you mind fixing those? Then I can merge since all else looks good. Thanks!

[radiangle]

@benshi34 we've renamed the trajectory files, could you review them again? thank you!