General Purpose Reranking With Instruction Following Capability

via “general-purpose reranking with instruction-following capability”

Domain-specific embedding models for RAG.

Unique: Reranking model with explicit instruction-following capability, enabling dynamic reranking behavior based on query intent or custom ranking criteria, beyond simple relevance scoring.

vs others: Outperforms Cohere rerank and Jina reranker on MTEB ranking benchmarks while supporting instruction-following for custom ranking logic, enabling more flexible and precise result ranking.

via “general instruction following and human preference alignment”

Alibaba's 32B reasoning model with chain-of-thought.

Unique: Uses a two-stage RL training approach where the second stage applies a general reward model and rule-based verifiers to align with human preferences across diverse tasks, enabling reasoning models to maintain instruction-following capability beyond specialized domains

vs others: Balances strong reasoning capability with general instruction-following through preference-aligned training, enabling use cases that require both transparent reasoning and practical task execution without requiring separate specialized models

via “instruction-following-capability-measurement”

* ⭐ 06/2022: [Solving Quantitative Reasoning Problems with Language Models (Minerva)](https://arxiv.org/abs/2206.14858)

Unique: BIG-bench treats instruction-following as a first-class capability measured across diverse task types rather than as a side effect of other capabilities, enabling researchers to isolate and study instruction-following as a distinct phenomenon

vs others: More comprehensive than instruction-following benchmarks focused on a single domain (e.g., code instruction-following) because it measures instruction-following across reasoning, knowledge, and language understanding tasks