Week of 2025-05-05 to 2025-05-12

Table of Contents

1. Weekly sampling of the week's emerging tech developments
2. Reasoning Models Don't Always Say What They Think
3. Llama.cpp Unifies and Extends Self-Hosted AI Vision Capabilities
4. Neuromorphic Computing Finding Application
5. Self Play and Improvement for LLMs
6. Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Models
7. Amazon Deploying Adaptive Arms
8. XR in the Real World

Weekly sampling of the week's emerging tech developments

• It feels like the first time in awhile, but there seems like there's been a relative lull in flashy AI news. That said, the pace of research continues at an incredible pace, with Huggingface's featured papers numbering 89 (to say nothing of the hundreds on Arxiv or in journals). Google IO may be a cause of and an end to the relative drought, with events starting Tuesday May 20th. All of this remains relative though, there has long been too much for anyone to keep up with. In any case, here's a sample including pieces on AI software, hardware, and theory along with robotics and XR.

Reasoning Models Don't Always Say What They Think

• https://arxiv.org/abs/2505.05410
• https://www.anthropic.com/research/reasoning-models-dont-say-think
• Depending on the benchmark, essentially all the leading AI models currently are 'reasoning' models that use a technique called 'chain of thought' (CoT) to improve their responses. While implementations differ, the technique aims to mitigate the limitations of autoregressive models, that is models that generate tokens strictly sequentially based on only the previous ones, by allowing a model to 'think' before giving a final response. Early on, this was conjectured to be likely useful for interpreting and debugging model responses, in particular looking for 'misalignment' where the model does something undesirable or subversive to user intent. Anthropic in this paper argues, at very least for Claude 3.7 and Deepseek R1, that 'chain of thought' is not capable of that. They set up a series of experiments where the model would be asked questions that it had 'hints' or 'reward hacks' for and analyzed the 'thinking' traces. Generally, there was no reliable indication that they referred to these implicitly illicit aids and appeared to construct fictional justifications to mask this. In other words, they don't 'say what they think'. Most of these reasoning models are heavily informed by reinforcement learning; these traces aren't 'reasoning' in an anthropomorphic sense but are rather an emergent behavior that optimizes problem solving accuracy, nothing more. Perhaps analogous to humans doing scratch work or sketches on an exam where there are fragments that help get to the answer but aren't high fidelity representations of 'thinking' itself.

Llama.cpp Unifies and Extends Self-Hosted AI Vision Capabilities

• https://github.com/ggml-org/llama.cpp/blob/master/docs/multimodal.md
• https://github.com/ggml-org/llama.cpp/releases/tag/b5331
• Llama.cpp, one of the leading open source self-hosted Large Language Model (LLM) runtimes, added unified support for some open weights models to process both text and visual input this week. Practically, this means most people can download an open and free model to a relatively recent laptop and run vision capable AI models on it. Originally there was support for the LLaVA (Large Language and Vision Assistant) architecture, however the complexities and unique architectures led to the creation of unique implementations for several models. Deemed unmaintainable, work began on 'mtmd' (MulTi-MoDal) as a universal replacement and was shipped in release b5331. Supported models include Gemma 3, SmolVLM 1 & 2, Pixtral 12B, Qwen 2 VL and Qwen 2.5 VL, Mistral Small 3.1 24B, among others.

Neuromorphic Computing Finding Application

• https://www.science.org/doi/10.1126/sciadv.adv2312
• A paper titled "Fully memristive spiking neural network for energy-efficient graph learning" was published in Science and claimed to experimentally demonstrate one of the first functional memristor based spiking neural networks. Memristors, or memory resistors, are relatively new to practical use but are fundamental components to circuit design, alongside resistors, capacitors, and inductors. In simple terms, they have a persistent but variable once set resistance that's dependent on the current previously passed through them. Spiking neural networks on the other hand take the classic neural network idea of emulating biological brains more literally by operating via discrete spikes as opposed to (practically) continuous transformations. Combined, these create a more direct analogue to biological neural networks and is part of the field of neuromorphic computing. The paper describes the manufacturing and use of such a memristive spiking neural network at non trivial scale and problem type. According to their results, the network was successfully applied to shortest path problem and graph classification use cases which have countless downstream applications, as simple as GPS navigation up to and beyond drug interaction study. Critically, even this experimental version consumed several orders of magnitude less energy than existing solutions. Replication has been difficult for this technology historically, but the benefits described and existing commercial interest make (relatively) speedy verification likely.

Self Play and Improvement for LLMs

• https://arxiv.org/abs/2505.03335
• https://github.com/LeapLabTHU/Absolute-Zero-Reasoner
• https://andrewzh112.github.io/absolute-zero-reasoner/
• Quality and quantity of training data has been a persistent struggle in AI, but perhaps not an eternal one according to this research. There have been various approaches to learning, but humans have ultimately been the ones assigning value or providing examples. This paper proposes and demonstrates an absolute zero reasoner training approach, wherein a model proposes, solves and judges in the training without human intervention. Alongside this the repository used to achieve strong results across many models was published, so replication should be swift and broad.

Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Models

• https://arxiv.org/abs/2505.04921
• We have here a useful systematic survey on the state of the art in reasoning models and provides a rough framework for understanding their capabilities. A dense but useful read.

Amazon Deploying Adaptive Arms

• https://www.aboutamazon.com/news/operations/amazon-vulcan-robot-pick-stow-touch
• https://manufacturingdigital.com/news/vulcan-amazons-first-ai-robot-that-can-see-and-feel
• https://techcrunch.com/2025/05/11/amazon-offers-peek-at-new-human-jobs-in-an-ai-bot-world/
• While certainly under development for years, Amazon recently publicized a robot they call 'Vulcan.' Unlike most existing industrial robotic arms, which are highly specialized and tightly programmed, this one uses a variety of fused sensors, various tools, and machine learning to more dynamically manipulate objects. According to Amazon these can already handle 75% of types of items in their warehouses. While somewhat vague, it does sound like these are being deployed in real roles in Amazon warehouses already, with broader rollout over the next 2 years.

XR in the Real World

• https://www.uploadvr.com/walking-70-miles-in-vision-pro-cubic-layer/
• https://apps.apple.com/us/app/cubiclayer/id6743873805
• Walk a mile in someone else's shoes, or walk 70 in an extended reality (XR) headset; at least one developer chose the latter. They built an app to take advantage of the environment mapping and mixed reality (MR) capabilities of the Apple Vision Pro and came away with a geospatial voxel editor. Or alternatively, a Minecraft inspired sandbox fixed in the real world. The videos in the UploadVR article are difficult to put to words and are worth the watch.