HF RL Explorer

I want embedding-backed passage rerankers for OpenDeepSearch.

I want embedding-backed passage rerankers for OpenDeepSearch.: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). The API should expose a BaseSemanticSearcher class whose public methods are calculate scores(queries: List str , documents: List str , normalize: str = "softmax") -…

The task

The API should expose a BaseSemanticSearcher class whose public methods are calculate_scores(queries: List[str], documents: List[str], normalize: str = "softmax") -> torch.Tensor, rerank(query: Union[str, List[str]], documents: List[str], top_k: int = 5, normalize: str =…

Part of XiaomiMiMo/MiMo-V2.6-RL-oss.