Decoupled Search Grounding (DSG) separates search functionality from reasoning models, enabling vendor-agnostic, tunable, and reusable search grounding. DSG achieves near-native accuracy on SimpleQA with 91% lower search cost and 99.4% warm-cache hit rate, while reducing latency by 68% and preserving concise output contracts.
arxiv
arXiv cs.AI
·
55d ago
·
research
Decoupling Search from Reasoning in LLM Agents
from English
Importance 3/3
New feature vs. leaders
New harness with differentiators
arXiv cs.AI
OpenAI
Anthropic
Google DeepMind
AI agents
Evaluation & benchmarks
Reasoning models
Related articles
arxiv
arXiv cs.CL
·
55d ago
·
2 views
Turing-RL: Learning User Simulators with Turing Rewards
Turing-RL introduces a reinforcement learning method using an LLM judge to evaluate how indistinguishable generated responses are from real user inputs. It outperforms baseline methods in both LLM and human evaluations across chat and Reddit forum domains, demonstrating that optimizing for indistinguishability improves user simulator performance.
arxiv
arXiv cs.CL
·
55d ago
·
1 view
Decoupling Search from Reasoning in LLM Agents
Decoupled Search Grounding (DSG) separates search functionality from reasoning models, enabling vendor-agnostic, tunable, and reusable search grounding. DSG achieves near-native accuracy on SimpleQA with 91% lower search cost and 99.4% warm-cache hit rate, while reducing latency by 68% and preserving concise output contracts.
arxiv
arXiv cs.CL
·
37d ago
GPT-5 outperforms humans in inducing belief states via planning
A new study evaluates Large Language Models' ability to induce specific belief states in other agents through actions rather than conversation, a capability termed Non-Conversational Planning ToM (NCP-ToM). Using the NCP-ExploreToM framework, researchers tested six frontier models and human participants on 600 task instances where agents had to move objects or direct characters to achieve belief goals.
arxiv
arXiv cs.AI
·
54d ago
·
6 views
NRT-Bench: Multi-turn Red-teaming of LLM Agents in Safety-Critical Systems
NRT-Bench introduces a benchmark for multi-turn red-teaming of LLM agents operating in a simulated nuclear power plant. Across four frontier operator models, 8.7% to 12.1% of attack sessions result in loss of a critical safety function, with vulnerabilities largely disjoint across models. The effectiveness of defences varies significantly by model, showing strong model dependence.
arxiv
arXiv cs.AI
·
54d ago
·
2 views
Defensive Misdirection Against Automated Attacks on Agentic AI
Agentic AI systems face growing threats from model-guided automated attacks. A new defense strategy, Contextual Misdirection via Progressive Engagement (CMPE), reduces attacker success rates by up to two orders of magnitude and nearly eliminates verified attack success in benchmark tests.