docs: improve SEO metadata across docs pages

docs/components/rerankers/config.mdx

@@ -1,6 +1,6 @@
 ---
 title: Config
-description: "Reference for shared and provider-specific reranker configuration options in Mem0, including top_k and API key settings."
+description: "Reference for Mem0 reranker configuration covering Cohere, Zero Entropy, sentence-transformer, HuggingFace, and LLM-based providers with top_k limits."
 ---
 
 ## Common Configuration Parameters

docs/components/rerankers/custom-prompts.mdx

@@ -1,6 +1,6 @@
 ---
 title: Custom Prompts
-description: "Customize the LLM reranker prompt template in Mem0 to control how search results are ranked and scored."
+description: "Write custom LLM reranker scoring prompts in Mem0 using query and document variables to tune relevance ranking for your specific domain or use case."
 ---
 
 When using LLM rerankers, you can customize the prompts used for ranking to better suit your specific use case and domain.

docs/components/rerankers/models/cohere.mdx

@@ -1,6 +1,6 @@
 ---
 title: Cohere
-description: "Configure Cohere as a reranker in Mem0 with support for English and multilingual reranking models."
+description: "Use Cohere rerank-v3.5 or older v3.0 models as a Mem0 reranker for English and multilingual search relevance with COHERE_API_KEY authentication."
 ---
 
 Cohere provides enterprise-grade reranking models with excellent multilingual support and production-ready performance.

docs/components/rerankers/models/huggingface.mdx

@@ -1,6 +1,6 @@
 ---
 title: Hugging Face Reranker
-description: 'Access thousands of reranking models from Hugging Face Hub'
+description: 'Access thousands of Hugging Face Hub reranking models in Mem0, including BAAI BGE cross-encoders, to boost memory search relevance and precision.'
 ---
 
 ## Overview

docs/components/rerankers/models/llm_reranker.mdx

@@ -1,6 +1,6 @@
 ---
 title: LLM Reranker
-description: 'Use any language model as a reranker with custom prompts'
+description: 'Use any Mem0-supported LLM as a reranker with custom prompts to score and rank memories by query relevance for flexible, tunable retrieval.'
 ---
 
 ## Overview

docs/components/rerankers/models/sentence_transformer.mdx

@@ -1,6 +1,6 @@
 ---
 title: Sentence Transformer
-description: 'Local reranking with HuggingFace cross-encoder models'
+description: 'Run local reranking in Mem0 with Sentence Transformers cross-encoder models like ms-marco-MiniLM for privacy-focused, on-premises deployments.'
 ---
 
 Sentence Transformer reranker provides local reranking using HuggingFace cross-encoder models, perfect for privacy-focused deployments where you want to keep data on-premises.

docs/components/rerankers/models/zero_entropy.mdx

@@ -1,6 +1,6 @@
 ---
 title: Zero Entropy
-description: "Configure Zero Entropy neural reranking models in Mem0 with zerank-1 and zerank-1-small support."
+description: "Configure Zero Entropy neural reranking in Mem0 with the zerank-1 and zerank-1-small models for fast, high-relevance search over stored memories."
 ---
 
 [Zero Entropy](https://www.zeroentropy.dev) provides neural reranking models that significantly improve search relevance with fast performance.

docs/components/rerankers/optimization.mdx

@@ -1,6 +1,6 @@
 ---
 title: Performance Optimization
-description: "Best practices for optimizing reranker performance in Mem0, covering candidate sizing, batching, and tuning."
+description: "Best practices for tuning reranker performance in Mem0, covering candidate set sizing, batching, model selection, and latency versus precision tradeoffs."
 ---
 
 Optimizing reranker performance is crucial for maintaining fast search response times while improving result quality. This guide covers best practices for different reranker types.

docs/components/rerankers/overview.mdx

@@ -1,6 +1,6 @@
 ---
 title: Overview
-description: 'Pick the right reranker path to boost Mem0 search relevance.'
+description: 'Compare Mem0 reranker providers including Cohere, Hugging Face, Sentence Transformers, and LLM-based rerankers to boost search relevance for agents.'
 ---
 
 Mem0 rerankers rescore vector search hits so your agents surface the most relevant memories. Use this hub to decide when reranking helps, configure a provider, and fine-tune performance.