7. Machine Bullshit This paper introduces the concept of machine bullshit, extending Harry Frankfurt’s definition, discourse made with indifference to truth, to LLMs.
RESEARCH
-

REST Benchmark Framework Evaluates Large Reasoning Models Robustness
By
–
5. Stress Testing Large Reasoning Models Proposes a new benchmark framework called REST to evaluate the robustness of Large Reasoning Models (LRMs) under multi-question stress.
-

Agentic-R1: 7B Language Model with Dynamic Tool-Based Reasoning
By
–
3. Agentic-R1 This paper introduces Agentic-R1, a 7B language model trained to dynamically switch between tool-based execution and pure text reasoning using a novel fine-tuning framework called DualDistill.
-

Chain-of-Thought Monitorability for AI Safety Oversight
By
–
4. Chain-of-Thought Monitorability Proposes that language-based CoT reasoning in LLMs offers an opportunity for AI safety by enabling automated oversight of models’ internal reasoning processes.
-

Study Evaluates LLM Performance Across Extended Context Lengths
By
–
2. Context Rot This comprehensive study by Chroma evaluates how state-of-the-art LLMs perform as input context length increases, challenging the common assumption that longer contexts are uniformly handled.
-

LLMs Performance on 2025 International Math Olympiad Evaluation
By
–
Not Even Bronze: Evaluating LLMs on 2025 International Math Olympiad https://
matharena.ai/imo/ Nice blog post from the team behind MathArena: Evaluating LLMs on Uncontaminated Math Competitions (
https://
arxiv.org/abs/2505.23281) providing independent analysis of LLM performance on IMO. -
Defining AI Agents: OpenAI Anthropic Disagreement
By
–
As always, that depends very much on much on how you (and they) define "agents" When not even OpenAI and Anthropic can settle on a shared definition between the two of them there's no hope for anyone else!
-

Meta-Agentic Master Key: Self-Evolving AGI Agent Factory
By
–
Meta-Agentic — The Master Key A self-evolving agent-factory that spins exponential design loops, harvests value across every vertical, and rewrites the rulebook as it learns. Demo (GitHub) : https://
github.com/MontrealAI/AGI
-Alpha-Agent-v0/tree/main/alpha_factory_v1/demos/meta_agentic_agi
… #AGI #AGIALPHA #ASIFirst -

ChatGPT Router Behavior Change Noticed
By
–

I feel like ChatGPT router already acts differently (dunno since when). Responses to non existing models (edge cases?) are handled by the model or router model with a slightly different system prompt. * model name changed for testing purposes only
-
LLM Capabilities Without Tools Exceed Expectations in Math
By
–
I thought that was true, but apparently I was wrong – LLMs without tools are a lot more capable than I expected, at least for things like IMO math problems I still think tools are the most important technique in AI engineering generally
