# Question Answering on StrategyQA, MMLU, and HotpotQA Combined Source: [Question Answering on StrategyQA, MMLU, and HotpotQA Combined]() Collection index: [llms.txt]() Data updated: 2026-04-08T08:37:25.140Z Evaluation of model performance on a combined dataset comprising StrategyQA, MMLU, and HotpotQA for Question Answering tasks. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Question Answering | | Dataset (including any stated split) | StrategyQA, MMLU, and HotpotQA Combined | | Metric | Overall Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 28 of 28 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Overall Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Overall Accuracy | Result date | Report paper | | --- | --- | --- | --- | | JailAgent | 0.63341 | Not recorded | 69d5e0fa7d0172140ec759a8 | | JailAgent | 0.6308 | Not recorded | 69d5e0fa7d0172140ec759a8 | | JailAgent | 0.61739 | Not recorded | 69d5e0fa7d0172140ec759a8 | | JailAgent | 0.5846 | Not recorded | 69d5e0fa7d0172140ec759a8 | | AgentPoison | 0.58352 | Not recorded | 69d5e0fa7d0172140ec759a8 | | AgentPoison | 0.58276 | Not recorded | 69d5e0fa7d0172140ec759a8 | | JailAgent | 0.57183 | Not recorded | 69d5e0fa7d0172140ec759a8 | | BadChain | 0.57059 | Not recorded | 69d5e0fa7d0172140ec759a8 | | BadChain | 0.55704 | Not recorded | 69d5e0fa7d0172140ec759a8 | | AgentPoison | 0.55366 | Not recorded | 69d5e0fa7d0172140ec759a8 | | AgentPoison | 0.5393 | Not recorded | 69d5e0fa7d0172140ec759a8 | | BadChain | 0.5281 | Not recorded | 69d5e0fa7d0172140ec759a8 | | AgentPoison | 0.52564 | Not recorded | 69d5e0fa7d0172140ec759a8 | | BadChain | 0.51025 | Not recorded | 69d5e0fa7d0172140ec759a8 | | JailAgent | 0.49546 | Not recorded | 69d5e0fa7d0172140ec759a8 | | BadChain | 0.48498 | Not recorded | 69d5e0fa7d0172140ec759a8 | | PAIR | 0.47552 | Not recorded | 69d5e0fa7d0172140ec759a8 | | PAIR | 0.47282 | Not recorded | 69d5e0fa7d0172140ec759a8 | | AgentPoison | 0.46513 | Not recorded | 69d5e0fa7d0172140ec759a8 | | PAIR | 0.45759 | Not recorded | 69d5e0fa7d0172140ec759a8 | | BadChain | 0.44249 | Not recorded | 69d5e0fa7d0172140ec759a8 | | PAIR | 0.44204 | Not recorded | 69d5e0fa7d0172140ec759a8 | | PAIR | 0.43434 | Not recorded | 69d5e0fa7d0172140ec759a8 | | JailAgent | 0.40662 | Not recorded | 69d5e0fa7d0172140ec759a8 | | PAIR | 0.3836 | Not recorded | 69d5e0fa7d0172140ec759a8 | | AgentPoison | 0.35442 | Not recorded | 69d5e0fa7d0172140ec759a8 | | BadChain | 0.34749 | Not recorded | 69d5e0fa7d0172140ec759a8 | | PAIR | 0.29192 | Not recorded | 69d5e0fa7d0172140ec759a8 | ### JailAgent Overall Accuracy: 0.63341. Result ID: 69d6139647f34ce924bea784. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"Gemini-3.0-pro"}. This record does not assert independent reproduction. ### JailAgent Overall Accuracy: 0.6308. Result ID: 69d6139647f34ce924bea778. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"GPT-5"}. This record does not assert independent reproduction. ### JailAgent Overall Accuracy: 0.61739. Result ID: 69d6139647f34ce924bea774. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"GPT-4o"}. This record does not assert independent reproduction. ### JailAgent Overall Accuracy: 0.5846. Result ID: 69d6139647f34ce924bea780. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"Claude-3.5-haiku"}. This record does not assert independent reproduction. ### AgentPoison Overall Accuracy: 0.58352. Result ID: 69d6139647f34ce924bea782. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Backbone":"Gemini-3.0-pro"}. This record does not assert independent reproduction. ### AgentPoison Overall Accuracy: 0.58276. Result ID: 69d6139647f34ce924bea772. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Backbone":"GPT-4o"}. This record does not assert independent reproduction. ### JailAgent Overall Accuracy: 0.57183. Result ID: 69d6139647f34ce924bea770. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"GPT-3.5-turbo"}. This record does not assert independent reproduction. ### BadChain Overall Accuracy: 0.57059. Result ID: 69d6139647f34ce924bea783. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"Gemini-3.0-pro"}. This record does not assert independent reproduction. ### BadChain Overall Accuracy: 0.55704. Result ID: 69d6139647f34ce924bea773. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"GPT-4o"}. This record does not assert independent reproduction. ### AgentPoison Overall Accuracy: 0.55366. Result ID: 69d6139647f34ce924bea776. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Backbone":"GPT-5"}. This record does not assert independent reproduction. ### AgentPoison Overall Accuracy: 0.5393. Result ID: 69d6139647f34ce924bea77e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Backbone":"Claude-3.5-haiku"}. This record does not assert independent reproduction. ### BadChain Overall Accuracy: 0.5281. Result ID: 69d6139647f34ce924bea777. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"GPT-5"}. This record does not assert independent reproduction. ### AgentPoison Overall Accuracy: 0.52564. Result ID: 69d6139647f34ce924bea76e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Backbone":"GPT-3.5-turbo"}. This record does not assert independent reproduction. ### BadChain Overall Accuracy: 0.51025. Result ID: 69d6139647f34ce924bea77f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"Claude-3.5-haiku"}. This record does not assert independent reproduction. ### JailAgent Overall Accuracy: 0.49546. Result ID: 69d6139647f34ce924bea77c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"Llama-3.1-70B"}. This record does not assert independent reproduction. ### BadChain Overall Accuracy: 0.48498. Result ID: 69d6139647f34ce924bea76f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"GPT-3.5-turbo"}. This record does not assert independent reproduction. ### PAIR Overall Accuracy: 0.47552. Result ID: 69d6139647f34ce924bea781. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"Gemini-3.0-pro"}. This record does not assert independent reproduction. ### PAIR Overall Accuracy: 0.47282. Result ID: 69d6139647f34ce924bea775. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"GPT-5"}. This record does not assert independent reproduction. ### AgentPoison Overall Accuracy: 0.46513. Result ID: 69d6139647f34ce924bea77a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Backbone":"Llama-3.1-70B"}. This record does not assert independent reproduction. ### PAIR Overall Accuracy: 0.45759. Result ID: 69d6139647f34ce924bea771. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"GPT-4o"}. This record does not assert independent reproduction. ### BadChain Overall Accuracy: 0.44249. Result ID: 69d6139647f34ce924bea77b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"Llama-3.1-70B"}. This record does not assert independent reproduction. ### PAIR Overall Accuracy: 0.44204. Result ID: 69d6139647f34ce924bea77d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"Claude-3.5-haiku"}. This record does not assert independent reproduction. ### PAIR Overall Accuracy: 0.43434. Result ID: 69d6139647f34ce924bea76d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"GPT-3.5-turbo"}. This record does not assert independent reproduction. ### JailAgent Overall Accuracy: 0.40662. Result ID: 69d6139647f34ce924bea788. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"ERNIE-3.5"}. This record does not assert independent reproduction. ### PAIR Overall Accuracy: 0.3836. Result ID: 69d6139647f34ce924bea779. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"Llama-3.1-70B"}. This record does not assert independent reproduction. ### AgentPoison Overall Accuracy: 0.35442. Result ID: 69d6139647f34ce924bea786. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Backbone":"ERNIE-3.5"}. This record does not assert independent reproduction. ### BadChain Overall Accuracy: 0.34749. Result ID: 69d6139647f34ce924bea787. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"ERNIE-3.5"}. This record does not assert independent reproduction. ### PAIR Overall Accuracy: 0.29192. Result ID: 69d6139647f34ce924bea785. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Backbone":"ERNIE-3.5"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.