Shalma Wegsman writes that physicists at MIT and Seoul National University have produced the first complete map of a crystal's quantum geometry, a hidden mathematical landscape that dictates how electrons behave. By combining measurements of Berry curvature and the quantum metric on a kagome solid, researchers have revealed a topological "ghost field" that influences electron motion. This new technique, which utilizes light to probe both the energy and velocity of electrons, offers a powerful tool for understanding exotic material properties and may aid in the search for room-temperature superconductors.
- The "quantum geometry" of a material consists of two parts: the Berry curvature (topological property) and the quantum metric (landscape steepness).
- The researchers used a kagome solid, a material with atoms arranged in a six-sided star pattern, for their primary measurement.
- A separate study by Bohm-Jung Yang's group applied the same method to black phosphorus, demonstrating the technique's versatility.
- The "ghost field" or "ghost charges" in topological materials cause electrons to move as if they are interacting with a force that does not physically exist.
- Understanding the quantum metric was recently crucial for explaining an exotic new form of superconductivity in a 2D crystal.
- The "quantum geometry" of a material consists of two parts: the Berry curvature (topological property) and the quantum metric (landscape steepness).
- The researchers used a kagome solid, a material with atoms arranged in a six-sided star pattern, for their primary measurement.
- A separate study by Bohm-Jung Yang's group applied the same method to black phosphorus, demonstrating the technique's versatility.
- The "ghost field" or "ghost charges" in topological materials cause electrons to move as if they are interacting with a force that does not physically exist.
- Understanding the quantum metric was recently crucial for explaining an exotic new form of superconductivity in a 2D crystal.
Jeff Shrager's repository introduces "executable archaeology" — the practice of recovering, transcribing, and running the original historical source of early computing programs so that the surviving artifact itself, rather than a modern rewrite, can serve as experimental evidence. The main project here is an IPL-V reimplementation of Victor Yngve's 1959–61 MIT sentence generator, run under Herbert A. Simon's account on 25 June 1962, whose surviving printout preserves the program, its phrase-structure grammar, execution traces, and handwritten corrections by Simon and his daughter Katherine.
- The repository also points to separate repos reconstructing the Logic Theorist in two historically distinct forms (LT1 in IPL-I and LT5 in IPL-V) and a full CTSS/IBM 7094 reanimation of Weizenbaum's original ELIZA
- A 2026 arXiv paper by Shrager and a 2025 IEEE Annals paper by Lane et al. provide the formal background
- Six stated principles govern the work, including distinguishing original behavior from behavior introduced by the reconstruction
- The repository also points to separate repos reconstructing the Logic Theorist in two historically distinct forms (LT1 in IPL-I and LT5 in IPL-V) and a full CTSS/IBM 7094 reanimation of Weizenbaum's original ELIZA
- A 2026 arXiv paper by Shrager and a 2025 IEEE Annals paper by Lane et al. provide the formal background
- Six stated principles govern the work, including distinguishing original behavior from behavior introduced by the reconstruction
Zhening Li and colleagues introduce JAZ, an LLM agent framework designed to minimize the complexity of agent loops by treating them as a programming language primitive called `invoke`. Instead of relying on external specialized systems for memory or self-improvement, JAZ enables agents to achieve these capabilities through code execution where all interactions are treated as variables within the environment. This minimalist approach allows highly expressive workflows, such as long-horizon recall and continual self-improvement, using only prompting rather than manually designed tools or complex architectures.
- The `invoke` primitive allows for recursive calls, enabling LLMs to write arbitrary executable code that includes further iterations of itself.
- In testing on the StuLife dataset, JAZ outperformed MemGPT (Letta) by 8% in recall performance while costing half as much.
- On self-improvement tasks using AppWorld, JAZ demonstrated a 4% improvement over ACE at a lower computational cost.
- The `invoke` primitive allows for recursive calls, enabling LLMs to write arbitrary executable code that includes further iterations of itself.
- In testing on the StuLife dataset, JAZ outperformed MemGPT (Letta) by 8% in recall performance while costing half as much.
- On self-improvement tasks using AppWorld, JAZ demonstrated a 4% improvement over ACE at a lower computational cost.
Milan Minsky writes that Leela AI transforms standard factory and warehouse cameras into smart sensors, offering an alternative to traditional IoT sensors by leveraging existing video feeds instead of physical hardware. The platform provides contextual visibility into operations, identifies bottlenecks, and tracks interactions between machines, operators, and materials without requiring retrofitting. It complements IoT systems by integrating with platforms like Velotic ThingWorx and AVEVA to create a comprehensive digital twin of manufacturing floors. The core technology utilizes MIT research-based AI, combining causal and neural networks for efficient data processing.
MIT scientists theorize in The Journal of Neuroscience that the brain generates cognition and consciousness through traveling waves of rhythmic neural activity performing analog computations. The brain uses local waves as a fast control system to coordinate neurons, with slower alpha/beta waves directing where/when faster gamma waves process sensory data.The authors argue that consciousness emerges when these wave patterns bring the cortex into a globally integrated state, and that the theory opens a path toward non-invasive, wave-based clinical treatments.
- Co-authors Scott Brincat and Jefferson Roy provided experimental evidence that waves influence neural spiking via ephaptic coupling (electric field-mediated interaction), not just synaptic transmission.
- Anesthesia studies with Emery N. Brown showed three structurally unrelated drugs all disrupt the same large-scale wave organization to induce unconsciousness, supporting a wave-based account over receptor-specific ones.
Mixed selectivity, where neurons respond to multiple cues, enhances brain computation but requires waves for coordination.
- Miller cautions the analog-computation claim is still theoretical and his lab's next step is to find direct signatures of wave interference in recorded brain-wave patterns.
- Co-authors Scott Brincat and Jefferson Roy provided experimental evidence that waves influence neural spiking via ephaptic coupling (electric field-mediated interaction), not just synaptic transmission.
- Anesthesia studies with Emery N. Brown showed three structurally unrelated drugs all disrupt the same large-scale wave organization to induce unconsciousness, supporting a wave-based account over receptor-specific ones.
Mixed selectivity, where neurons respond to multiple cues, enhances brain computation but requires waves for coordination.
- Miller cautions the analog-computation claim is still theoretical and his lab's next step is to find direct signatures of wave interference in recorded brain-wave patterns.
Deploying Large Language Models in streaming applications is limited by growing KV cache memory during decoding and an inability to generalize beyond training sequence length, with naive window attention failing once text exceeds the cache size. The authors observe an attention sink phenomenon where models assign strong attention scores to initial tokens even when semantically irrelevant, and find that retaining the KV of those initial sinks together with a sliding window recovers performance. This motivates StreamingLLM, a zero-shot framework that enables LLMs trained on finite windows to generalize to infinite sequence lengths without fine-tuning, achieving stable language modeling up to 4 million tokens on Llama-2, MPT, Falcon and Pythia.
- Attention sinks arise from strong attention to initial tokens acting as a sink for excess attention mass.
- A placeholder token added as a dedicated attention sink during pre-training further improves streaming deployment.
- StreamingLLM achieves up to 22.2x speedup over sliding-window recomputation in streaming settings.
- Paper is ICLR 2024 and code/datasets are released at mit-han-lab/streaming-llm.
- Attention sinks arise from strong attention to initial tokens acting as a sink for excess attention mass.
- A placeholder token added as a dedicated attention sink during pre-training further improves streaming deployment.
- StreamingLLM achieves up to 22.2x speedup over sliding-window recomputation in streaming settings.
- Paper is ICLR 2024 and code/datasets are released at mit-han-lab/streaming-llm.
A research collaboration between a U.S. Air Force cadet and an MIT Lincoln Laboratory researcher explored whether nontechnical service members can develop software via vibe-coding—using prompts to guide generative chatbots in writing code. The study revealed that while large language models are excellent prototyping tools for communicating user needs, they present significant challenges regarding security, accuracy, and the need for rigorous human review when handling sensitive data.
* Capability of nonexperts to create functional application prototypes
* Challenges in scaling from complex tactical uses toward practical document processing tasks
* Security risks associated with unintended data transmission during model interaction
* Capability of nonexperts to create functional application prototypes
* Challenges in scaling from complex tactical uses toward practical document processing tasks
* Security risks associated with unintended data transmission during model interaction
As organizations increasingly integrate artificial intelligence into hiring processes and performance reviews, experts suggest that true AI fluency is shifting from technical tool mastery toward a capacity for human judgment. Rather than simply learning specific software interfaces, employees must develop a conceptual framework to understand where AI systems are reliable versus risky.
Key insights include:
- Treating AI outputs as iterative drafts rather than final answers to ensure accuracy and accountability.
- Addressing the inequality of access regarding paid tools, high-speed internet, and time for experimentation.
- Avoiding "performative checkbox" exercises that focus on usage metrics like prompt counts instead of actual business outcomes.
- Protecting against assessment bias to ensure neurodivergent employees are not disadvantaged by specific communication styles during fluency evaluations.
Key insights include:
- Treating AI outputs as iterative drafts rather than final answers to ensure accuracy and accountability.
- Addressing the inequality of access regarding paid tools, high-speed internet, and time for experimentation.
- Avoiding "performative checkbox" exercises that focus on usage metrics like prompt counts instead of actual business outcomes.
- Protecting against assessment bias to ensure neurodivergent employees are not disadvantaged by specific communication styles during fluency evaluations.
Nobel laureate Daron Acemoglu critiques current optimism regarding AI productivity and economic narratives. He argues that much of the prevailing debate is speculative and fails to address critical issues like concentrated corporate power and extractive data models. Rather than focusing on whether capitalism is mutating, he suggests evaluating technology based on whether it fosters inclusive or extractive institutions.
>*Seen through that lens, AI is not troublesome in its own right, but rather whether it is positioned as inclusive or extractive. Today’s AI hyperscalers, he argues, fit the extractive mold almost perfectly: concentrated ownership, regulatory capture, and a business model that extracts data and attention at scale."
- Skepticism toward massive near-term AI productivity gains due to current model limitations
- The distinction between simple automation and true human-complementary tasks
- Potential social instability if significant job displacement occurs among younger generations
- A call for global governance and a focus on socially desirable technological outcomes
>*Seen through that lens, AI is not troublesome in its own right, but rather whether it is positioned as inclusive or extractive. Today’s AI hyperscalers, he argues, fit the extractive mold almost perfectly: concentrated ownership, regulatory capture, and a business model that extracts data and attention at scale."
- Skepticism toward massive near-term AI productivity gains due to current model limitations
- The distinction between simple automation and true human-complementary tasks
- Potential social instability if significant job displacement occurs among younger generations
- A call for global governance and a focus on socially desirable technological outcomes
* **Problem:** LLMs struggle to derive reliable meaning from raw sensor signals, often producing non-actionable or factually incorrect interpretations of time-series data.
* **Methodology:** The study implements a structured RAG-based prompt structure that combines water consumption measurements with descriptive statistics and qualitative user information (such as household water practices).
* **Key Finding:** Augmenting prompts with multidimensional contextual information leads to much higher evaluation scores for grounding, pattern recognition, and actionable recommendations.
* **Methodology:** The study implements a structured RAG-based prompt structure that combines water consumption measurements with descriptive statistics and qualitative user information (such as household water practices).
* **Key Finding:** Augmenting prompts with multidimensional contextual information leads to much higher evaluation scores for grounding, pattern recognition, and actionable recommendations.
