AI Wiki — Knowledge Base

Hi everyone, I'm building a local AI agent with Ollama and exploring dynamic PDF extraction. Since Ollama can't directly process PDFs, I'm extracting text and passing it via prompts. Should I use PDFPlumber, a vector database (RAG), or another approach for accurate document understanding? Guide me !

User seeks guidance on PDF processing methods for local AI agents built with Ollama, comparing approaches like PDFPlumber extraction, vector database RAG systems, and alternative techniques for accura

local-air-ollama

8 Jun 2026

Search: “retrieval”

From Statute to Control Flow: Span-Grounded Deontic Trees for Defeasible Scope Parsing

Generalized Rank-based Evaluation for Knowledge Graph Completion: Perspectives, Framework, and Analyses

Generative Reasoning Re-ranker

HDRAgent: An Agentic Framework for Multi-Exposure HDR Imaging

How Much Dense Attention is Necessary? Oracle-Guided Sparse Prefill for Full/GQA Layers in Hybrid Long-Context Models

KITE: A Tri-Modal Transformer Integrating Text, Images, and Knowledge Graphs for Fake News Detection

MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation

MemoryVLA++: Temporal Modeling via Memory and Imagination in Vision-Language-Action Models

MemoVAD: Resource-Efficient Video Anomaly Detection via Dynamic Semantic Memory in Edge Computing Scenarios

MotionVLA: Injecting Geometric Motion into Vision-Language-Action Model

Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs

PhysGraph: A Physics-aware 3D Scene Graph for Perception and Reasoning

Physics-Guided Dual Decoding and Spectral Supervision for Global 3D Hydrometeor Prediction

PRISM: Recovering Instruction Sets from Language Model Activations

Remember with Confidence: Uncertainty Quantification for Spatio-temporal Memory with Probabilistic Guarantees

Report: GKE Inference Gateway delivers up to 92% faster AI responses

TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking

TempoBench: Evaluating Temporal Causal Reasoning in Large Language Models

Testing the Black Box: Structural Barriers to Independent Evaluation of Consumer-Facing Health LLMs

The Injection Paradox: Brand-Level Suppression in Safety-Trained LLM Recommendations via RAG Context Injection

TQA-Bench: Evaluating LLMs for Multi-Table Question Answering

UAOR: Uncertainty-aware Observation Reinjection for Vision-Language-Action Models

Understanding Benchmark Language Under Weakened Formal Semantics

ViMax: Agentic Video Generation

Why Limit the Residual Stream to Layers and Not Tokens? Persistent Memory for Continuous Latent Reasoning

Zero-Shot Semantic Re-Identification for Autonomous Driving: A VLM Baseline Study

AdMem: Advanced Memory for Task-solving Agents

CAPE: Contrastive Action-conditioned Parallel Encoding for Embodied Planning

ChemQuests: A Curated Chemistry Question-Answer Database Extracted from ChemRxiv papers

Decision-Aware Evaluation of Physics-Informed Surrogates

Diagnosing Visual Ignorance in Vision-Language Models

GOPAgen: Motion-Aware and Efficient Agentic Long-Video Understanding with Structural Memory and Hierarchical Reasoning

Just-In-Time Reinforcement Learning: Continual Learning in LLM Agents Without Gradient Updates

M^3Exam: Benchmarking Multimodal Memory for Realistic User-Agent Interactions

Meaning in Order, Order in Meaning: Semantic R-precision for Keyphrase Evaluation

MHA-RAG: Improving Efficiency, Accuracy, and Consistency by Encoding Exemplars as Soft Prompts

Mind the Gap: Bridging Behavioral Silos with LLMs in Multi-Vertical Recommendations

Principles of Concept Representation in Sentence Encoders

Rethinking Genomic Modeling Through Optical Character Recognition

SERNF: Sample-Efficient Real-World Dexterous Policy Fine-Tuning via Action-Chunked Critics and Normalizing Flows

Sparse Subspace-to-Expert Sharing for Task-Agnostic Continual Learning

StainFlow: Entity-Stain Tracking and Evidence Linking for Process Rewards in GUI Agents

Style or Content? Evaluating Style Classifiers with Controlled Content Overlap

TEVI: Text-Conditioned Editing of Visual Representations via Sparse Autoencoders for Improved Vision-Language Alignment

Trace Reconstruction with Language Models

Trading Engagement for Sustainability: Carbon-Aware Re-ranking for E-commerce Recommendations

Tree-of-Experience: A Structured Experience-Management Solution for Self-Evolving Agents under Low-Repetition and Implicit-Reward Environments