If you can just trivially create synthetic data from one language to the other, then that's not a new concept of a programming language, just a remapping. No need to retrain the LLM, just use a tool-calling LLM.
PROMPT ENGINEERING
-
Jules by Google to get Memory feature soon
By
–
Jules by Google is about to get Memory soon! As well as a new file selector in the prompt composer. "Enable memories to let Jules use context from your past tasks to improve its responses"
-
Creating Concise Languages That Map to Popular Targets
By
–
The best you can do for now is create a less verbose language that maps onto a widely used target language. You can then generate paired training examples programmatically so the LLM learns to operate with fewer tokens. (Congratulations, you also just reinvented Triton.)
-
The Challenge of Evaluating LLMs: Balancing Marketing Claims and Independent Testing
By
–
It’s kind of a dilemma. You want to check independent evals because the original ones might be inflated for marketing purposes.
At the same time, independent evals may also undersell the LLM because of accidental bad prompting, bad batching, bad optimization etc -
LFM2-Extract Model Recommended for Extraction Tasks
By
–
They haven't been fine-tuned for this precise purpose, but I'd recommend checking LFM2-Extract (1.2B or 350M).
-
Multi-Model Agents and High-Dimensional Capability Optimization
By
–
multi model is a big part of the Agent Labs thesis
— swyx 🐣 (@swyx) 26 septembre 2025
havent published findings but yeah you kinda want to argmax harness(capability)
so if you believe that capability is a high dimensional space (as in, “good at coding” is not enough, it breaks down to a hundred subtasks of… https://t.co/8yO4HdtbjFmulti model is a big part of the Agent Labs thesis havent published findings but yeah you kinda want to argmax harness(capability) so if you believe that capability is a high dimensional space (as in, “good at coding” is not enough, it breaks down to a hundred subtasks of
-
Perplexity Tests New ‘Sonar Testing’ Reasoning Model
By
–
BREAKING 🚨: Perplexity is testing a new "Sonar Testing" reasoning model internally. Potentially, it will arrive as a reasoning upgrade to the existing Sonar model.
— 🚨 AI News | TestingCatalog (@testingcatalog) 26 septembre 2025
Imo, "Sonar Testing" is the best name ever 👀 pic.twitter.com/a1ChGPDGLhBREAKING : Perplexity is testing a new "Sonar Testing" reasoning model internally. Potentially, it will arrive as a reasoning upgrade to the existing Sonar model. Imo, "Sonar Testing" is the best name ever
-
Clarifying Prompt Injection vs Jailbreaking Misconceptions
By
–
Plus I'd had limited success getting people to understand "prompt injection" because too many people assumed it meant people "injecting bad prompts" into the system themselves, conflating it with jailbreaking
-
Lethal Trifecta: Data Exfiltration Through Prompt Injection Attacks
By
–
That's regular prompt injection. I came up with the "lethal trifecta" framing to help people understand the data exfiltration subset of prompt injection, because the same exact bug kept on showing up again and again and again across different systems