AI Wiki — Knowledge Base

【採用情報】Member of Technical Staff (LLM Development) の募集を開始しました🐟 https://sakana.ai/careers/member-of-technical-staff-llm-development/ Sakana N…

【採用情報】Member of Technical Staff (LLM Development) の募集を開始しました🐟 https://sakana.ai/careers/member-of-technical-staff-llm-development/ Sakana Namazuを起点に拡大しているSakana AIのLLM開発を担うポジションです。下記のような挑戦が待っています。・事前

researchdavid-ha--x

21 Aug 2026

Search: “llm”

FIDES: A Concordance Protocol for LLM-Generated Trading Strategies

Forgetting to Forget: Attention Sink as A Gateway for Backdooring LLM Unlearning

FormuEvo: LLM-Guided Evolution for Discovering Solver-Efficient Mixed-Integer Programming Formulations

How do you go from jailbreaking your PlayStation to CEO of Canada's leading LLM provider? You go to the University of Toronto. 'When you go …

InjecMEM: Memory Injection Attack on LLM Agent Memory Systems

Kernel Token Contradiction: a Fast and Principled Approach for LLM Claim Uncertainty Quantification

LLM assisted writing deserves empirical evaluation

LLM-based Agents for Forecasting and Prediction: Methods, Training, Evaluation, and Applications

MemGuard: Persisting Verifier Signals for LLM-Agent Memory Governance

Most of the LLM routing gap is task type

NeST: Neighborhood-aware semantic alignment and temporal modulation for LLM based time series forecasting

NoTB: Oracle-Free Triage of LLM-Generated RTL via Cross-Model Formal Consensus

PropUQ-MAS: Propagation-Aware Uncertainty Quantification for LLM Multi-Agent Systems

Register Shifts Break LLM Safety: A Bengali Benchmark with Culturally Grounded Harms

Restore LLM Inference Capacity in Seconds with Shadow Engine Recovery in NVIDIA Dynamo

Right-Sizing LLM-Agent Decomposition in VAT Determination: A Pilot Controlled Sweep

STONIC: A Layered Measurement Contract for LLM Value Profiling

When Not to Imitate: Boundary-Aware Skill Memory for Reliable Tool-Use LLM Agents

Affective Context Amplifies Sycophancy in LLM Responses

aiXamine: Unified Black-Box Evaluation of Cross-Dimensional Trade-offs in LLM Safety, Security, and Privacy

Causal Modeling of Adverse Pregnancy Outcomes via Adaptive LLM Proposals

ClawSentry: A Progressive Multi-Tier Security Monitor for Safeguarding Autonomous LLM Agents

is there a music UPSCALER llm?

Reinforcing Multi-Turn Reasoning in LLM Agents via Fine-Grained Reward Structure and Credit Assignment

New/Old benchmark that provides a lot of answers for local LLM

HYDRA: A Heterogeneous Chiplet DSE Framework for Serving Dynamic Hybrid LLM Workloads

【採用情報】Member of Technical Staff (LLM Development) の募集を開始しました🐟 https://sakana.ai/careers/member-of-technical-staff-llm-development/ Sakana N…

MileGPO: Milestone Inference with Local Evidence for Graph-Based Policy Optimization of Long-Horizon LLM Agents

Optimal Skill Selection for LLM Agents with Provable Bicriteria Guarantees

RepSelect: Robust LLM Unlearning via Representation Selectivity

“The key is symbolic AI layered in with an LLM. @GaryMarcus was right all along”

Which Eviction Policy Should an LLM Cache Use? A Systematic Study Across Workloads, Capacities, and Encoders

Flama: a Python framework for development and deployment of production-ready APIs, machine learning, and LLM services

home lab assistant llm

LLM-Powered Predictive Decision-Making for Sustainable Data Center Operations

Pedagogical AI in Mental Health: A Tri-Stream Fine-Tuned LLM Framework for Automated Clinical Supervision and Risk Triage

TrojanGYM: A Detector-in-the-Loop LLM for Adaptive RTL Hardware Trojan Insertion

An Empirical Study of Reward Specification and Benchmark Reliability in GRPO-based LLM Unlearning

Beyond BFI: The CSI for Enhanced Reliability and Validity in Evaluating LLM Personality Traits

OraclePhys: A Systematic Framework for LLM Fine-Tuning on Structural Mechanics

Task-Aware Harness Provisioning for LLM Agents in Mission-Critical Infrastructure Operations

TSQueryBench: LLM-as-a-Judge for Time Series Explanations

A Scalable Pipeline for LLM-Teacher Distillation Labeling: Work-Stealing Job Scheduling and Memory-Aware GPU Concurrency

AeroCopilotBench: A Two-Tier Benchmark for Evaluating LLM Agents as Aviation Copilots in an Interactive Virtual Cockpit Environment

Agentic-SQL Revisited: Autonomy-Based Taxonomy and Empirical Benchmark Analysis for LLM Text-to-SQL

'Agents = model + harness + context' is a much better mental model than treating the LLM as the entire product. The model provides capabilit…

From LLM Inference to Agentic Workloads: Characterization and Implications for Serving Systems

HyperSkill: Self-Evolving LLM Agents via Hypergraph-Structured Skill Memory