Introducing the File Search Tool in Gemini API https://
buff.ly/PI21mdd
#AI #MachineLearning #DeepLearning #LLMs #DataScience
LLMS
-
Gemini API Introduces File Search Tool for LLMs
By
–
-
RouteLLM Routes Prompts to Optimal AI Models on ChatLLM
By
–
RouteLLM is available on ChatLLM and routes to the best model based on your prompt We use the following models – GPT 5.2
– Sonnet 4.5
– Opus 4.5
– Gemini 3.0 Flash
– Grok 4.1 We combine them all to give you the best experience -
OpenAI develops ChatGPT Jobs for AI career guidance
By
–
BREAKING : OpenAI is developing ChatGPT Jobs, a new career-focused AI agent designed to help users with resume tips, job search, and career guidance. AI Job search will be a big market too
-
LLM-enabled Multi-Agent Systems Survey for AI Developers
By
–
LLM-enabled Multi-Agent Systems Nice little survey on the progress in multi-agent systems. Great resources for AI devs to catch up on agents.
-
New System Reduces AI Refusal Rates with Minimal Compute
By
–
Because the system harnesses internal activations already happening within a model, and reserves heavier computation only for potentially harmful exchanges, it adds only ~1% compute overhead. It’s also more accurate, with an 87% drop in refusal rates on harmless requests.
-
Claude’s Interpretability Probe Screens Traffic via Internal Activations
By
–
Our new system adds several innovations. One is a practical application of interpretability: a probe that can see Claude’s internal activations helps to screen all traffic. These activations are like Claude’s gut instincts, and they’re harder to fool.
-
Claude’s Classifiers Reduce Jailbreak Success Rate to 4.4%
By
–
The classifiers reduced the jailbreak success rate from 86% to 4.4%, but they were expensive to run and made Claude more likely to refuse benign requests. We also found the system was still vulnerable to two types of attacks, shown in the figure below:
-
Anthropic’s Constitutional Classifiers Advance Jailbreak Protection
By
–
New Anthropic Research: next generation Constitutional Classifiers to protect against jailbreaks. We used novel methods, including practical application of our interpretability work, to make jailbreak protection more effective—and less costly—than ever.
-
GPT-5.2 Pro Achieves Milestone on Erdos Mathematical Problems
By
–
Milestone in AI tools applied to Erdos problems, with GPT-5.2 Pro:
-
AI21 Labs Shares Research on Scaling Agentic SWE-bench Evaluation
By
–
Our Research team just dropped a few behind-the-scenes blogs on scaling agentic SWE-bench evaluation, including the failure modes we hit and what finally worked. I'm curious to hear your thoughts about our work