Tootfinder

Multi-Hop Question Answering: When Can Humans Help, and Where do They Struggle?
Jinyan Su, Claire Cardie, Jennifer Healey
https://arxiv.org/abs/2510.04493 https://

Multi-Hop Question Answering: When Can Humans Help, and Where do They Struggle?
Multi-hop question answering is a challenging task for both large language models (LLMs) and humans, as it requires recognizing when multi-hop reasoning is needed, followed by reading comprehension, logical reasoning, and knowledge integration. To better understand how humans might collaborate effectively with AI, we evaluate the performance of crowd workers on these individual reasoning subtasks. We find that while humans excel at knowledge integration (97\% accuracy), they often fail to recog…

Microsoft AI CEO Mustafa Suleyman says Microsoft plans to focus on superintelligence that prioritizes human control; he will lead a new superintelligence team (Reed Albergotti/Semafor)
https://www.semafor.com/article/11/05/2025/mi…

Microsoft superintelligence team promises to keep humans in charge
Microsoft AI’s “humanist” approach to superintelligence will be led by Mustafa Suleyman.

Collaboration and Conflict between Humans and Language Models through the Lens of Game Theory
Mukul Singh, Arjun Radhakrishna, Sumit Gulwani
https://arxiv.org/abs/2509.04847 htt…

Collaboration and Conflict between Humans and Language Models through the Lens of Game Theory
Language models are increasingly deployed in interactive online environments, from personal chat assistants to domain-specific agents, raising questions about their cooperative and competitive behavior in multi-party settings. While prior work has examined language model decision-making in isolated or short-term game-theoretic contexts, these studies often neglect long-horizon interactions, human-model collaboration, and the evolution of behavioral patterns over time. In this paper, we investig…

I see that with a lot of criticism of generative "AI"—people state that obviously it's completely unreliable and untrustworthy for _their domain of expertise_ but they'll somehow gladly use it for other stuff.
I believe this cognitive dissonance has to do with how the chatbots pretend to be humans and trick us to assume agency when there is none.
Anyway, as I said otherwise it's great, you should read it: https://theoatmeal.com/comics/ai_art

Re the “wise sage” option:
You do not raise money to build multiple new nuclear power plants to power your product if people think your pitch is “it sucks just like humans except faster and more unpredictably, and it’s worse at arithmetic”

Real humans don’t stream Drake songs 23 hours a day, rapper suing Spotify says - Ars Technica
https://arstechnica.com/tech-policy/2025/11/real-humans-dont-stream-drake-songs-23-hours-a-day-rapper-suing-spotify-says/

Real humans don’t stream Drake songs 23 hours a day, rapper suing Spotify says
Proposed class action may force Spotify to pay back artists harmed by streaming fraud.

Pointing-Guided Target Estimation via Transformer-Based Attention
Luca M\"uller, Hassan Ali, Philipp Allgeuer, Luk\'a\v{s} Gajdo\v{s}ech, Stefan Wermter
https://arxiv.org/abs/2509.05031

Pointing-Guided Target Estimation via Transformer-Based Attention
Deictic gestures, like pointing, are a fundamental form of non-verbal communication, enabling humans to direct attention to specific objects or locations. This capability is essential in Human-Robot Interaction (HRI), where robots should be able to predict human intent and anticipate appropriate responses. In this work, we propose the Multi-Modality Inter-TransFormer (MM-ITF), a modular architecture to predict objects in a controlled tabletop scenario with the NICOL robot, where humans indicate…

Sturgeon’s law remains my biggest reassurance against LLM world takeover. As far as I can tell, that law is as solid as gravity. Humans on average not giving a shit IS the training material poison pill.
yes, I have just looked at random GitHub code again, why do you ask

"I want all social discourse to fit within my personal Overton Window." -- Basically all humans

"We stand by the accuracy of what we published,"
Reuters' statement read.
"We have carefully reviewed the published footage, and we have found no reason to believe Reuters longstanding commitment to accurate, unbiased journalism has been compromised."
The four-minute clip, released Sept. 3, captured Putin telling Xi that biotechnology could one day extend human life indefinitely.
"With continuous advances in biotechnology, human organs wi…

Reuters withdraws Putin-Xi immortality video after Chinese legal objections
Reuters on Sept. 5 withdrew a video in which Russian President Vladimir Putin and Chinese President Xi Jinping were shown discussing whether humans could live to 150 years old, after China's state broadcaster revoked permission to use the footage.

crabs have evolved five times, humans (at least the ones alive today) only once. how dare we describe ourselves as the peak of evolution?

OH on slack:
SREcon 2024: hey, just a crazy thing, but we thought about giving the root password to a parrot. What do you think?
SREcon 2025: here's our comprehensive report from parrot's rampage, after equipping it with root password and a spending account
SREcon 2026: codebases should be parrots, not pets
SREcon 2027, presented mostly by parrots: how to work effectively with grumpy humans
#sre #devoops #aipocalypse @sre@tagpush.app

The Raising Humans Kind Podcast
Great Australian Pods Podcast Directory: #GreatAusPods

From Actions to Kinesics: Extracting Human Psychological States through Bodily Movements
Cheyu Lin, Katherine A. Flanigan
https://arxiv.org/abs/2510.04844 https://

From Actions to Kinesics: Extracting Human Psychological States through Bodily Movements
Understanding the dynamic relationship between humans and the built environment is a key challenge in disciplines ranging from environmental psychology to reinforcement learning (RL). A central obstacle in modeling these interactions is the inability to capture human psychological states in a way that is both generalizable and privacy preserving. Traditional methods rely on theoretical models or questionnaires, which are limited in scope, static, and labor intensive. We present a kinesics recog…

LLMs as Policy-Agnostic Teammates: A Case Study in Human Proxy Design for Heterogeneous Agent Teams
Aju Ani Justus, Chris Baber
https://arxiv.org/abs/2510.06151 https://

LLMs as Policy-Agnostic Teammates: A Case Study in Human Proxy Design for Heterogeneous Agent Teams
A critical challenge in modelling Heterogeneous-Agent Teams is training agents to collaborate with teammates whose policies are inaccessible or non-stationary, such as humans. Traditional approaches rely on expensive human-in-the-loop data, which limits scalability. We propose using Large Language Models (LLMs) as policy-agnostic human proxies to generate synthetic data that mimics human decision-making. To evaluate this, we conduct three experiments in a grid-world capture game inspired by Sta…

Picture the human body. Zoom in on a single cell. It lives for a while, then splits or dies, as part of a community of cells that make up a particular tissue. This community lives together for many many cell-lifetimes, each performing their own favorite function and reproducing as much as necessary to maintain their community, consuming the essential resources they need and contributing back what they can so that the whole body can live for decades. Each community of cells is interdependent on the whole body, but also stable and sustainable over long periods of time.
Now imagine a cancer cell. It has lost its ability to harmonize with the whole and prioritize balance, instead consuming and reproducing as quickly as it can. As neighboring tissues start to die from its excess, it metastasizes, always spreading to new territory to fuel its unbalanced appetite. The inevitable result is death of the whole body, although through birth, that body can create a new fresh branch of tissues that may continue their stable existence free of cancer. Alternatively, radiation or chemotherapy might be able to kill off the cancer, at great cost to the other tissues, but permitting long-term survival.
To the cancer cell, the idea of decades-long survival of a tissue community is unbelievable. When your natural state is unbounded consumption, growth, and competition, the idea of interdependent cooperation (with tissues all around the body you're not even touching, no less) seems impossible, and the idea that a tissue might survive in a stable form for decades is ludicrous.
"Perhaps if conditions were bleak enough to perfectly balance incessant unrestrained growth against the depredations of a hostile environment it might be possible? I guess the past must have been horribly brutal, so that despite each tissue trying to grow as much as possible they each barely survived? Yes, a stable and sustainable population is probably only possible under conditions of perfectly extreme hardship, and in our current era of unfettered growth, we should rejoice that we live in much easier times!"
You can probably already see where I'm going with this metaphor, but did you know that there are human communities, alive today, that have been living sustainably for *tens, if not hundreds of thousands of years*?
#anarchy #colonialism #civilization
P.S. if you're someone who likes to think about past populations and historical population growth, I cannot recommend the (short, free) game Opera Omnia by Stephen Lavelle enough: https://www.increpare.com/2009/02/opera-omnia/

AP2O: Correcting LLM-Generated Code Errors Type by Type Like Humans via Adaptive Progressive Preference Optimization
Jianqing Zhang, Wei Xia, Hande Dong, Qiang Lin, Jian Cao
https://arxiv.org/abs/2510.02393

AP2O: Correcting LLM-Generated Code Errors Type by Type Like Humans via Adaptive Progressive Preference Optimization
LLMs' code generation capabilities have yielded substantial improvements in the effectiveness of programming tasks. However, LLM-generated code still suffers from compilation and runtime errors. Existing offline preference optimization methods primarily focus on enhancing LLMs' coding abilities using pass/fail signals in the preference data, overlooking the deep-level error types in the failed codes. To address this, we propose Adaptively Progressive Preference Optimization (AP2O) for coding (i…

When AI Gets Persuaded, Humans Follow: Inducing the Conformity Effect in Persuasive Dialogue
Rikuo Sasaki, Michimasa Inaba
https://arxiv.org/abs/2510.04229 https://

When AI Gets Persuaded, Humans Follow: Inducing the Conformity Effect in Persuasive Dialogue
Recent advancements in AI have highlighted its application in captology, the field of using computers as persuasive technologies. We hypothesized that the "conformity effect," where individuals align with others' actions, also occurs with AI agents. This study verifies this hypothesis by introducing a "Persuadee Agent" that is persuaded alongside a human participant in a three-party persuasive dialogue with a Persuader Agent. We conducted a text-based dialogue experiment with human participants…

In an alternate universe, there are marsupial humans.

A profile of Mercor, a Scale AI rival valued at $2B in February, which hires domain experts to train models; it had a $100M run rate in March and $6M H1 profit (Richard Nieva/Forbes)
https://www.forbes.com/sites/richardnieva/2025/09/03/ai…

AI’s Next Job? Recruiting People To Train More AI
Mercor, debuting on the Forbes Cloud 100 list, built an AI recruiter to interview job candidates. Then it discovered a cash cow: finding humans to train AI models.

LLMs are a fundamentally useless technology because their applications (supposedly) boil down to humans not having to think for themselves or do their own writing / drawing / filming.
But if you can do it on your own - why would you need a robot to do it? It’s, at best, a novelty.
That’s why this shit only resonates with executives and capital owners. “Get things done with fewer people and expenses” is at least an actual pitch. “Get things done faster for yourself” isn’t.
The individual angle really works for things you already were trying to avoid doing because you’re either disinterested or don’t have enough time to do things right.
“Avoid your work” as a value proposition doesn’t work when you’re dealing with intellectual labor rather than commodities. Not large scale, not long term.
Sorry for the rant, I saw some Notion ads on the subway and got irritated 😅
#AI #llm #LLMs

Here's my latest Human Meme podcast! We cry because we can!
#human

Curiosity-Driven Co-Development of Action and Language in Robots Through Self-Exploration
Theodore Jerome Tinker, Kenji Doya, Jun Tani
https://arxiv.org/abs/2510.05013 https://

Curiosity-Driven Co-Development of Action and Language in Robots Through Self-Exploration
Human infants acquire language and action co-developmentally, achieving remarkable generalization capabilities from only a minimal number of learning examples. In contrast, recent large language models require exposure to billions of training tokens to achieve such generalization. What mechanisms underlie such efficient developmental learning in humans? This study addresses this question through simulation experiments in which robots learn to perform various actions corresponding to imperative …

Happy to see resolution for this case of anti-health, anti-vax, quackery and grifting.
I believe most farmers are ethical and understand that disease within their flocks and herds can and do cause disease in other domestic and wild populations and those diseases are often what causes pandemics in humans.
Avian flu is a danger. Millions of birds have died because of it, and have been culled on other farms to try to control it.
Ostriches are no different.
https://www.cbc.ca/news/canada/british-columbia/livestory/bc-ostrich-farm-decision-scoc-9.6968394

Advanced 2.5 Million-Year-Old Tools May Rewrite Human History https://www.404media.co/advanced-2-5-million-year-old-tools-may-rewrite-human-history/

«Lessons in humility & simplicity for 'data science': Garmin's health status»
I blogged about how another #wearable manufacturer went down the road of leaving up data interpretation to humans instead of automating it – and how that relates to "AI" or "automated decision making"
(Responses to this toot become blog comments too)
#quantifiedself #personalscience

🦾 Humans sense a collaborating robot as part of their 'extended' body
#robots

Humans sense a collaborating robot as part of their 'extended' body
Researchers from the Istituto Italiano di Tecnologia (IIT) in Genoa (Italy) and Brown University in Providence (U.S.) have discovered that people sense the hand of a humanoid robot as part of their body schema, particularly when it comes to carrying out a task together, like slicing a bar of soap.

Training a Perceptual Model for Evaluating Auditory Similarity in Music Adversarial Attack
Yuxuan Liu, Rui Sang, Peihong Zhang, Zhixin Li, Shengchen Li
https://arxiv.org/abs/2509.04985

Training a Perceptual Model for Evaluating Auditory Similarity in Music Adversarial Attack
Music Information Retrieval (MIR) systems are highly vulnerable to adversarial attacks that are often imperceptible to humans, primarily due to a misalignment between model feature spaces and human auditory perception. Existing defenses and perceptual metrics frequently fail to adequately capture these auditory nuances, a limitation supported by our initial listening tests showing low correlation between common metrics and human judgments. To bridge this gap, we introduce Perceptually-Aligned M…

Leech (2022) by Hiron Ennes is a post-post-apoc novel told from the POV of a hive-mind parasite that's eliminated any human who has medical knowledge, except for the humans it inhabits. This is rationalized as a survival tactic.
The story begins with a mystery: the parasite has no memory of the death of its host who was doctor to a tyrant in the bleak north, so it sends another of itself to investigate. It discovers the dead host has a different parasite. A dog eats that. Some peop…

A Case for Declarative LLM-friendly Interfaces for Improved Efficiency of Computer-Use Agents
Yuan Wang, Mingyu Li, Haibo Chen
https://arxiv.org/abs/2510.04607 https://

A Case for Declarative LLM-friendly Interfaces for Improved Efficiency of Computer-Use Agents
Computer-use agents (CUAs) powered by large language models (LLMs) have emerged as a promising approach to automating computer tasks, yet they struggle with graphical user interfaces (GUIs). GUIs, designed for humans, force LLMs to decompose high-level goals into lengthy, error-prone sequences of fine-grained actions, resulting in low success rates and an excessive number of LLM calls. We propose Goal-Oriented Interface (GOI), a novel abstraction that transforms existing GUIs into three decla…

Replaced article(s) found for cs.NE. https://arxiv.org/list/cs.NE/new
[1/1]:
- The dynamic interplay between in-context and in-weight learning in humans and neural networks
Jacob Russin, Ellie Pavlick, Michael J. Frank

Sources: Google Gemini co-lead Noam Shazeer clashed with colleagues on internal forums over topics like gender and Gaza; moderators deleted some of his comments (The Information)
https://www.theinformation.com/articles/googl…

Google’s $2.7 Billion AI Hire Tests Company’s Speech Limits With Inflammatory Posts
This spring, Google AI leader Noam Shazeer posted a comment in one of the company’s internal discussion forums, in response to a post about how Google employees could support their transgender and nonbinary colleagues on International Transgender Day of Visibility. “I do not believe that humans ...

Biosphere Substrate and its parameters range
Yegor A. Morozov, Mikhail Bukhtoyarov, Mahdi Yoozbashizadeh
https://arxiv.org/abs/2509.04846 https://arxiv.org…

Biosphere Substrate and its parameters range
The term Biosphere Substrate is introduced for celestial bodies, like terrestrial planets or big gas giant moons, suitable to sustain full-scale open biosphere after Terraforming. The purpose of the work is to examine the range of parameters for a biosphere substrate. Most importantly, Biosphere Substrate gravity should sustain full-scale open breathable atmosphere dense enough after Terraforming during the lifecycle of Planetary System. Humans can not reproduce in too low gravity, can not rest…

ProToM: Promoting Prosocial Behaviour via Theory of Mind-Informed Feedback
Matteo Bortoletto, Yichao Zhou, Lance Ying, Tianmin Shu, Andreas Bulling
https://arxiv.org/abs/2509.05091

ProToM: Promoting Prosocial Behaviour via Theory of Mind-Informed Feedback
While humans are inherently social creatures, the challenge of identifying when and how to assist and collaborate with others - particularly when pursuing independent goals - can hinder cooperation. To address this challenge, we aim to develop an AI system that provides useful feedback to promote prosocial behaviour - actions that benefit others, even when not directly aligned with one's own goals. We introduce ProToM, a Theory of Mind-informed facilitator that promotes prosocial actions in mul…

Emergent Social Dynamics of LLM Agents in the El Farol Bar Problem
Ryosuke Takata, Atsushi Masumori, Takashi Ikegammi
https://arxiv.org/abs/2509.04537 https://

Emergent Social Dynamics of LLM Agents in the El Farol Bar Problem
We investigate the emergent social dynamics of Large Language Model (LLM) agents in a spatially extended El Farol Bar problem, observing how they autonomously navigate this classic social dilemma. As a result, the LLM agents generated a spontaneous motivation to go to the bar and changed their decision making by becoming a collective. We also observed that the LLM agents did not solve the problem completely, but rather behaved more like humans. These findings reveal a complex interplay between …

An Arbitration Control for an Ensemble of Diversified DQN variants in Continual Reinforcement Learning
Wonseo Jang, Dongjae Kim
https://arxiv.org/abs/2509.04815 https://

An Arbitration Control for an Ensemble of Diversified DQN variants in Continual Reinforcement Learning
Deep reinforcement learning (RL) models, despite their efficiency in learning an optimal policy in static environments, easily loses previously learned knowledge (i.e., catastrophic forgetting). It leads RL models to poor performance in continual reinforcement learning (CRL) scenarios. To address this, we present an arbitration control mechanism over an ensemble of RL agents. It is motivated by and closely aligned with how humans make decisions in a CRL context using an arbitration control of m…

"UNESCO Adds an Area the Size of Bolivia to Biosphere Reserve System to Protect 5% of the World’s Land"
#UNESCO #Environment #Conservation

UNESCO Adds an Area the Size of Bolivia to Reserves That Protect 5% of the World's Land
The Reserves are governed in a way that Man and the biosphere he needs to survive can do so for as long as we humans live on this planet.

A Hybrid CAPTCHA Combining Generative AI with Keystroke Dynamics for Enhanced Bot Detection
Ayda Aghaei Nia
https://arxiv.org/abs/2510.02374 https://arxiv.…

A Hybrid CAPTCHA Combining Generative AI with Keystroke Dynamics for Enhanced Bot Detection
Completely Automated Public Turing tests to tell Computers and Humans Apart (CAPTCHAs) are a foundational component of web security, yet traditional implementations suffer from a trade-off between usability and resilience against AI-powered bots. This paper introduces a novel hybrid CAPTCHA system that synergizes the cognitive challenges posed by Large Language Models (LLMs) with the behavioral biometric analysis of keystroke dynamics. Our approach generates dynamic, unpredictable questions tha…

@… Unpopular, or just unpopular among humans? 😅

Humans are such a moronic species. How we survived this long is beyond me.
(The points made in this little animation have been backed up by substantial research. This is just a cute presentation of it.)
https://youtu.be/Omc37TvHN74

AD-NODE: Adaptive Dynamics Learning with Neural ODEs for Mobile Robots Control
Shao-Yi Yu, Jen-Wei Wang, Maya Horii, Vikas Garg, Tarek Zohdi
https://arxiv.org/abs/2510.05443 htt…

AD-NODE: Adaptive Dynamics Learning with Neural ODEs for Mobile Robots Control
Mobile robots, such as ground vehicles and quadrotors, are becoming increasingly important in various fields, from logistics to agriculture, where they automate processes in environments that are difficult to access for humans. However, to perform effectively in uncertain environments using model-based controllers, these systems require dynamics models capable of responding to environmental variations, especially when direct access to environmental information is limited. To enable such adaptiv…

Leopards eating faces party is a time-honored tradition. #science

Early Humans Weren’t Apex Predators—They Were Eaten by Leopards
Homo habilis was thought to be the first hominin to use stone tools for hunting and processing meat, but they might have been prey instead of predators.

What Types of Code Review Comments Do Developers Most Frequently Resolve?
Saul Goldman, Hong Yi Lin, Jirat Pasuksmit, Patanamon Thongtanunam, Kla Tantithamthavorn, Zhe Wang, Ray Zhang, Ali Behnaz, Fan Jiang, Michael Siers, Ryan Jiang, Mike Buller, Minwoo Jeong, Ming Wu
https://arxiv.org/abs/2510.05450

What Types of Code Review Comments Do Developers Most Frequently Resolve?
Large language model (LLM)-powered code review automation tools have been introduced to generate code review comments. However, not all generated comments will drive code changes. Understanding what types of generated review comments are likely to trigger code changes is crucial for identifying those that are actionable. In this paper, we set out to investigate (1) the types of review comments written by humans and LLMs, and (2) the types of generated comments that are most frequently resolved …

@… First I've heard of this - very exciting! https://www.cbc.ca/news/science/jeremy-hansen-moon-1.7649455

In 4 months, Canadian Jeremy Hansen will blast off to the moon — where he'll also be a human guinea pig | CBC News
If all goes to plan, next February could see the return of humans to the moon. Four people — NASA astronauts Victor Glover, Reid Wiseman, Christina Koch and Canadian Space Agency astronaut Jeremy Hansen — will see it as no one has before. And they will also be collecting valuable health data on themselves.

Har brukt Qobuz i noen måneder nå, og anbefaler den. Jeg bruker ikke strŸmming (og hadde ikke Spotify fŸr) men ville ha en tjeneste hvor jeg kan kjŸpe og laste ned digitale musikkfiler i toppkvalitet og vite at jeg beholder dem selv om de forsvinner fra tjenesten (*host* Apple Music *host*). Qobuz innfrir, og en ekstra bonus er at de som abonnerer på strŸmmetjenesten får en betydelig rabatt på de albumene man kjŸper.
Dersom du ikke vil ha strŸmming, men heller vil at ekte mennesker lag…

An eclectic musical adventure...
Radio Paradise streams highly curated, eclectic mixes of music -- chosen by real humans -- with unparalleled audio quality.

The human biological advantage over AI
William Stewart
https://arxiv.org/abs/2509.04130 https://arxiv.org/pdf/2509.04130

The human biological advantage over AI
Recent advances in AI raise the possibility that AI systems will one day be able to do anything humans can do, only better. If artificial general intelligence (AGI) is achieved, AI systems may be able to understand, reason, problem solve, create, and evolve at a level and speed that humans will increasingly be unable to match, or even understand. These possibilities raise a natural question as to whether AI will eventually become superior to humans, a successor "digital species", with a rightfu…

Some Further Developments on a Neurobiologically-based Model for Color Sensations in Humans
Charles Q. Wu
https://arxiv.org/abs/2510.01000 https://arxiv.or…

Some Further Developments on a Neurobiologically-based Model for Color Sensations in Humans
At HVEI-2012, I presented a neurobiologically-based model for trichromatic color sensations in humans, mapping the neural substrate for color sensations to V1-L4: the thalamic recipient layer of the primary visual cortex. In this paper, I propose that V1-L4 itself consists of three distinct sub-layers that directly correspond to the three primary color sensations: blue, red, and green. Furthermore, I apply this model to three aspects of color vision: the three-dimensional (3D) color solid, dich…

The Interplay of Attention and Memory in Visual Enumeration
B. Sankar, Devottama Sen, Dibakar Sen
https://arxiv.org/abs/2510.05833 https://arxiv.org/pdf/25…

The Interplay of Attention and Memory in Visual Enumeration
Humans navigate and understand complex visual environments by subconsciously quantifying what they see, a process known as visual enumeration. However, traditional studies using flat screens fail to capture the cognitive dynamics of this process over the large visual fields of real-world scenes. To address this gap, we developed an immersive virtual reality system with integrated eye-tracking to investigate the interplay between attention and memory during complex enumeration. We conducted a tw…

Multi-faceted light pollution modelling and its application to the decline of artificial illuminance in France
Rolf Buhler, Philippe Deverch\`ere, Christophe Plotard, S\'ebastien Vauclair
https://arxiv.org/abs/2510.02977

Multi-faceted light pollution modelling and its application to the decline of artificial illuminance in France
Artificial Light At Night (ALAN) has been increasing steadily over the past century, particularly during the last decade. This leads to rising light pollution, which is known to have adverse effects on living organisms, including humans. We present a new software package to model light pollution from ground radiance measurements. The software is called Otus 3 and incorporates innovative ALAN diffusion models with different atmospheric profiles, cloud covers and urban emission functions. To date…

How Does Cognitive Bias Affect Large Language Models? A Case Study on the Anchoring Effect in Price Negotiation Simulations
Yoshiki Takenami, Yin Jou Huang, Yugo Murawaki, Chenhui Chu
https://arxiv.org/abs/2508.21137

How Does Cognitive Bias Affect Large Language Models? A Case Study on the Anchoring Effect in Price Negotiation Simulations
Cognitive biases, well-studied in humans, can also be observed in LLMs, affecting their reliability in real-world applications. This paper investigates the anchoring effect in LLM-driven price negotiations. To this end, we instructed seller LLM agents to apply the anchoring effect and evaluated negotiations using not only an objective metric but also a subjective metric. Experimental results show that LLMs are influenced by the anchoring effect like humans. Additionally, we investigated the rel…

Mask2IV: Interaction-Centric Video Generation via Mask Trajectories
Gen Li, Bo Zhao, Jianfei Yang, Laura Sevilla-Lara
https://arxiv.org/abs/2510.03135 https://

Mask2IV: Interaction-Centric Video Generation via Mask Trajectories
Generating interaction-centric videos, such as those depicting humans or robots interacting with objects, is crucial for embodied intelligence, as they provide rich and diverse visual priors for robot learning, manipulation policy training, and affordance reasoning. However, existing methods often struggle to model such complex and dynamic interactions. While recent studies show that masks can serve as effective control signals and enhance generation quality, obtaining dense and precise mask an…

Spooky month is upon us 👻

"Chimpanzees revise their beliefs if they encounter new information, a hallmark of rationality that was once assumed to be unique to humans"
(Original title: Chimps Are Capable of Human-Like Rational Thought, Breakthrough Study Finds)
https://www.404med…

Shape and word parts combine linearly in the Bouba–Kiki effect https://link.springer.com/article/10.3758/s13414-025-03151-1

Shape and word parts combine linearly in the Bouba–Kiki effect - Attention, Perception, & Psychophysics
Languages have evolved in part due to cross-modal associations between shape and sound. A famous example is the Bouba–Kiki effect, wherein humans associate words like bouba/kiki to round/angular shapes. How does the Bouba–Kiki effect work for natural words and shapes that contain a mixture of features? If the effect is holistic, the effect for a composite stimulus would not be explainable using the parts. If the effect is compositional, it will be. Here we provide evidence for the latter po…

In Ukraine, Trump's "Peace Efforts" result in even more savage attacks on civilians by Russia.
Those who see supporting Ukraine as a waste of resources are cheering as innocent lives are ruined and ended prematurely.
Perhaps even more disgusting and shameful than the genocide Russia wages on Ukraine, is the indifference shown by so many people around the world.
Monsters and ghouls who insist they are decent humans.
Giorgio is a true journalist reporting…

Resisting Indifference: How Pokovsk Is Not the Gateway to Donbas
Live from Ukraine 🇺🇦

Sources: ex-xAI researcher Eric Zelikman is raising $1B at a $5B valuation for Humans&, which aims to train AI that is better at collaborating with humans (Anna Tong/Forbes)
https://www.forbes.com/sites/annatong/2025…

xAI Researcher In Talks To Raise $1 Billion For New Frontier Lab Humans&
Founded by former xAI staffer Eric Zelikman and early Googler Georges Harik, the new lab is building models that are better at interacting with humans.

#ChristopherHitchens

This week's Mindscape podcast is with Petter Törnberg who has been running agent based simulated social networks to try and figure out why they're so divisive and polarizing.
His models do apparently capture these features of the networks which promote the most popular posts to everyone.
Sad to learn that simply using chronological feeds apparently not only doesn't help he reckons it makes things worse!
Maybe his agent's just aren't smart enough to follow the right people, but then probably nor are mere humans.
🤔
#socialMedia #podcast #fediverse

KSON: A 💌 to the humans maintaining computer configurations
KSON combines the strengths of YAML and JSON. With KSON you can write files that are easy to read, edit, and debug—without worrying about invisible whitespace errors or missing commas.
🛠️ https://kson.org

🌝 Solar powered moon brick factory could build future lunar cities
#space

Solar powered moon brick factory could build future lunar cities
Chinese scientists have developed a remarkable machine that could revolutionize how humans build structures on the moon. The device works like a 3D printer powered by concentrated sunlight, turning lunar soil (known as regolith) into strong construction bricks without needing any materials from Earth.

In his book Mathematica David Bessis notes this curios fact:
The _definition_ of a 'species' in biology is a bit like an equivalence relation in math: Two animals are of the same species if they can produce fertile offspring. The species are the equivalence classes.
But then, biologists are much more practically inclined to work with their definitions, or how do they know that elephants and humans are in fact distinct species? How do they check?
https://en.wikipedia.org/wiki/Species

“Thanks to the solidity of bearded vulture nest structures and their locations in the western Mediterranean… they have acted as natural museums, conserving historical material in good condition,” the researchers wrote
Among the centuries’ worth of eggshells, prey remains, and natural nesting material, researchers identified 226 objects that were either made or altered by humans.
These included weaponry like a crossbow bolt and wooden lance,
decorated sheep leather,
…

Behind the Blog: Making Things for Humans https://www.404media.co/behind-the-blog-making-things-for-humans/

there are 500 trillion calories of humans in the world 😋

This is near where I live and I'm livid about the sheer waste of money to create something so disruptive to both animals and humans. https://www.theguardian.com/us-news/2025/oct/01/arizona-border-wall-san-rafael-valley

MM-Nav: Multi-View VLA Model for Robust Visual Navigation via Multi-Expert Learning
Tianyu Xu, Jiawei Chen, Jiazhao Zhang, Wenyao Zhang, Zekun Qi, Minghan Li, Zhizheng Zhang, He Wang
https://arxiv.org/abs/2510.03142

MM-Nav: Multi-View VLA Model for Robust Visual Navigation via Multi-Expert Learning
Visual navigation policy is widely regarded as a promising direction, as it mimics humans by using egocentric visual observations for navigation. However, optical information of visual observations is difficult to be explicitly modeled like LiDAR point clouds or depth maps, which subsequently requires intelligent models and large-scale data. To this end, we propose to leverage the intelligence of the Vision-Language-Action (VLA) model to learn diverse navigation capabilities from synthetic expe…

Research: AI's ability to complete long and complex software engineering tasks doubles every 6-7 months, but there is a "messiness tax" for real-world tasks (Boaz Barak/Windows On Theory)
https://windowsontheory.org/2025/11/04/thoughts-by…

Thoughts by a non-economist on AI and economics
Crossposted on lesswrong Modern humans first emerged about 100,000 years ago. For the next 99,800 years or so, nothing happened. Well, not quite nothing. There were wars, political intrigue, the in…

Can AI really replace humans when it comes to the arts? https://www.theguardian.com/commentisfree/picture/2025/oct/11/can-ai-really-replace-humans-when-it-comes-to-the-arts?CMP=Share_iOSApp_Other…

Desperate companies now hiring humans to fix what #AI botched https://futurism.com/companies-hiring-humans-fix-ai I similarly wouldn't be surprised about an upcoming

Desperate Companies Now Hiring Humans to Fix What AI Botched
After getting rid of their human workers to invest in AI, companies are now hiring freelancers to fix the sloppy content those AIs put out.

As humans, we •all• live our lives with our heads up our butts — including but not limited to the Britannica writers. If we’re lucky, someone will help us see what we’re missing. If we’re wise, we listen. At its best, that’s what Wikipedia can do — and in public, at scale, for everyone.
7/

"Study busts big bad myth that wolves are growing fearless of humans"
#Wolves #Animals
https:…

Study busts big bad myth that wolves are growing fearless of humans
After years of being hunting to near-extinction, wolves are making a comeback in some of their historical ranges in Europe and North America. With their return comes a growing concern that the predators are moving too close to human settlements, becoming bolder and less fearful of people. Researchers set out to see if that’s true, […]

Long-Term Human Motion Prediction Using Spatio-Temporal Maps of Dynamics
Yufei Zhu, Andrey Rudenko, Tomasz P. Kucner, Achim J. Lilienthal, Martin Magnusson
https://arxiv.org/abs/2510.03031

Long-Term Human Motion Prediction Using Spatio-Temporal Maps of Dynamics
Long-term human motion prediction (LHMP) is important for the safe and efficient operation of autonomous robots and vehicles in environments shared with humans. Accurate predictions are important for applications including motion planning, tracking, human-robot interaction, and safety monitoring. In this paper, we exploit Maps of Dynamics (MoDs), which encode spatial or spatio-temporal motion patterns as environment features, to achieve LHMP for horizons of up to 60 seconds. We propose an MoD-i…

Towards Cognitively-Faithful Decision-Making Models to Improve AI Alignment
Cyrus Cousins, Vijay Keswani, Vincent Conitzer, Hoda Heidari, Jana Schaich Borg, Walter Sinnott-Armstrong
https://arxiv.org/abs/2509.04445

Towards Cognitively-Faithful Decision-Making Models to Improve AI Alignment
Recent AI work trends towards incorporating human-centric objectives, with the explicit goal of aligning AI models to personal preferences and societal values. Using standard preference elicitation methods, researchers and practitioners build models of human decisions and judgments, which are then used to align AI behavior with that of humans. However, models commonly used in such elicitation processes often do not capture the true cognitive processes of human decision making, such as when peop…

How do Humans and LLMs Process Confusing Code?
Youssef Abdelsalam, Norman Peitek, Anna-Maria Maurer, Mariya Toneva, Sven Apel
https://arxiv.org/abs/2508.18547 https://

How do Humans and LLMs Process Confusing Code?
Already today, humans and programming assistants based on large language models (LLMs) collaborate in everyday programming tasks. Clearly, a misalignment between how LLMs and programmers comprehend code can lead to misunderstandings, inefficiencies, low code quality, and bugs. A key question in this space is whether humans and LLMs are confused by the same kind of code. This would not only guide our choices of integrating LLMs in software engineering workflows, but also inform about possible …

#RIP Jane Goodall.
https://apnews.com/article/jane-goodall-obit-78698397851bc7634717206f7eba07b2

Jane Goodall, pioneering conservationist and primatologist, dies at 91
Jane Goodall, the conservationist renowned for her groundbreaking chimpanzee field research and globe-spanning environmental advocacy, has died. She was 91. The Jane Goodall Institute announced her death Wednesday, saying she died of natural causes while in California on a U.S. speaking tour. Goodall is best remembered for living among chimpanzees in Africa decades ago and documenting the animals using tools and doing other activities previously believed to be exclusive to humans. She also note…

InfoMosaic-Bench: Evaluating Multi-Source Information Seeking in Tool-Augmented Agents
Yaxin Du, Yuanshuo Zhang, Xiyuan Yang, Yifan Zhou, Cheng Wang, Gongyi Zou, Xianghe Pang, Wenhao Wang, Menglan Chen, Shuo Tang, Zhiyu Li, Siheng Chen
https://arxiv.org/abs/2510.02271

InfoMosaic-Bench: Evaluating Multi-Source Information Seeking in Tool-Augmented Agents
Information seeking is a fundamental requirement for humans. However, existing LLM agents rely heavily on open-web search, which exposes two fundamental weaknesses: online content is noisy and unreliable, and many real-world tasks require precise, domain-specific knowledge unavailable from the web. The emergence of the Model Context Protocol (MCP) now allows agents to interface with thousands of specialized tools, seemingly resolving this limitation. Yet it remains unclear whether agents can ef…

OpenAI releases GDPval, a benchmark to test AI performance on "economically valuable, real-world tasks", and says Claude Opus 4.1 was the best performing model (Maxwell Zeff/TechCrunch)
https://techcrunch.com/2025/09/25/openai-says-g…

OpenAI says GPT-5 stacks up to humans in a wide range of jobs | TechCrunch
A new test from OpenAI aims to understand how close AI is to outperforming humans at economically valuable work.

📏 Learning from punishment: Model makes sense of the cognitive processes humans use
#psychology

Learning from punishment: Model makes sense of the cognitive processes humans use
From toddlers' timeouts to criminals' prison sentences, punishment reinforces social norms, making it known that an offender has done something unacceptable. At least, that is usually the intent—but the strategy can backfire. When a punishment is perceived as too harsh, observers can be left with the impression that an authority figure is motivated by something other than justice.

Oruga: An Avatar of Representational Systems Theory
Daniel Raggi, Gem Stapleton, Mateja Jamnik, Aaron Stockdill, Grecia Garcia Garcia, Peter C-H. Cheng
https://arxiv.org/abs/2509.04041

Oruga: An Avatar of Representational Systems Theory
Humans use representations flexibly. We draw diagrams, change representations and exploit creative analogies across different domains. We want to harness this kind of power and endow machines with it to make them more compatible with human use. Previously we developed Representational Systems Theory (RST) to study the structure and transformations of representations. In this paper we present Oruga (caterpillar in Spanish; a symbol of transformation), an implementation of various aspects of RST.…

Virtual Community: An Open World for Humans, Robots, and Society
Qinhong Zhou, Hongxin Zhang, Xiangye Lin, Zheyuan Zhang, Yutian Chen, Wenjun Liu, Zunzhe Zhang, Sunli Chen, Lixing Fang, Qiushi Lyu, Xinyu Sun, Jincheng Yang, Zeyuan Wang, Bao Chi Dang, Zhehuan Chen, Daksha Ladia, Jiageng Liu, Chuang Gan
https://arxiv.org/abs/2508.14893

Virtual Community: An Open World for Humans, Robots, and Society
The rapid progress in AI and Robotics may lead to a profound societal transformation, as humans and robots begin to coexist within shared communities, introducing both opportunities and challenges. To explore this future, we present Virtual Community-an open-world platform for humans, robots, and society-built on a universal physics engine and grounded in real-world 3D scenes. With Virtual Community, we aim to study embodied social intelligence at scale: 1) How robots can intelligently cooperat…

In-Context Learning can Perform Continual Learning Like Humans
Liuwang Kang, Fan Wang, Shaoshan Liu, Hung-Chyun Chou, Chuan Lin, Ning Ding
https://arxiv.org/abs/2509.22764 https…

In-Context Learning can Perform Continual Learning Like Humans
Large language models (LLMs) can adapt to new tasks via in-context learning (ICL) without parameter updates, making them powerful learning engines for fast adaptation. While extensive research has examined ICL as a few-shot learner, whether it can achieve long-term retention and cross-task knowledge accumulation when multitasks arrive sequentially remains underexplored. Motivated by human memory studies, we investigate the retention characteristics of ICL in multitask settings and extend it to …

Analyzing Reluctance to Ask for Help When Cooperating With Robots: Insights to Integrate Artificial Agents in HRC
Ane San Martin, Michael Hagenow, Julie Shah, Johan Kildal, Elena Lazkano
https://arxiv.org/abs/2509.01450

Analyzing Reluctance to Ask for Help When Cooperating With Robots: Insights to Integrate Artificial Agents in HRC
As robot technology advances, collaboration between humans and robots will become more prevalent in industrial tasks. When humans run into issues in such scenarios, a likely future involves relying on artificial agents or robots for aid. This study identifies key aspects for the design of future user-assisting agents. We analyze quantitative and qualitative data from a user study examining the impact of on-demand assistance received from a remote human in a human-robot collaboration (HRC) assem…

"Why a wildfire chemical toxic to humans lingers longer in clouds"
#Wildfires #Chemicals
https://

Why a wildfire chemical toxic to humans lingers longer in clouds
Researchers from the University of Bristol and the Central Laser Facility (CLF) at the Science and Technology Facilities Council (STFC) have discovered why a toxic compound found in wildfire smoke resists natural breakdown when dissolved in atmospheric water droplets.

SpaceX defends lunar lander after challenge from NASA administrator
In a lengthy blog post Thursday, Elon Musk’s space company said it “rapidly advanced” the core Starship spacecraft through 11 test flights
and reached 49 milestones in developing the HLS (Human Landing System) version of the craft designed to reach the lunar surface as part of the government’s Artemis program.
“Starship continues to simultaneously be the fastest path to returning humans to the surface of t…

Analysis of Bluffing by DQN and CFR in Leduc Hold'em Poker
Tarik Zaciragic, Aske Plaat, K. Joost Batenburg
https://arxiv.org/abs/2509.04125 https://arx…

Analysis of Bluffing by DQN and CFR in Leduc Hold'em Poker
In the game of poker, being unpredictable, or bluffing, is an essential skill. When humans play poker, they bluff. However, most works on computer-poker focus on performance metrics such as win rates, while bluffing is overlooked. In this paper we study whether two popular algorithms, DQN (based on reinforcement learning) and CFR (based on game theory), exhibit bluffing behavior in Leduc Hold'em, a simplified version of poker. We designed an experiment where we let the DQN and CFR agent play ag…

A study focused on OpenAI's GPT-4o mini found that LLMs can be persuaded to comply with objectionable requests using the same tactics that persuade humans (Dina Bass/Bloomberg)

Learning Human-Perceived Fakeness in AI-Generated Videos via Multimodal LLMs
Xingyu Fu, Siyi Liu, Yinuo Xu, Pan Lu, Guangqiuse Hu, Tianbo Yang, Taran Anantasagar, Christopher Shen, Yikai Mao, Yuanzhe Liu, Keyush Shah, Chung Un Lee, Yejin Choi, James Zou, Dan Roth, Chris Callison-Burch
https://arxiv.org/abs/2509.22646

Learning Human-Perceived Fakeness in AI-Generated Videos via Multimodal LLMs
Can humans identify AI-generated (fake) videos and provide grounded reasons? While video generation models have advanced rapidly, a critical dimension -- whether humans can detect deepfake traces within a generated video, i.e., spatiotemporal grounded visual artifacts that reveal a video as machine generated -- has been largely overlooked. We introduce DeeptraceReward, the first fine-grained, spatially- and temporally- aware benchmark that annotates human-perceived fake traces for video generat…

Beyond Words: Interjection Classification for Improved Human-Computer Interaction
Yaniv Goren, Yuval Cohen, Alexander Apartsin, Yehudit Aperstein
https://arxiv.org/abs/2509.03181

Beyond Words: Interjection Classification for Improved Human-Computer Interaction
In the realm of human-computer interaction, fostering a natural dialogue between humans and machines is paramount. A key, often overlooked, component of this dialogue is the use of interjections such as "mmm" and "hmm". Despite their frequent use to express agreement, hesitation, or requests for information, these interjections are typically dismissed as "non-words" by Automatic Speech Recognition (ASR) engines. Addressing this gap, we introduce a novel task dedicated to interjection classifica…

AgentPack: A Dataset of Code Changes, Co-Authored by Agents and Humans
Yangtian Zi, Zixuan Wu, Aleksander Boruch-Gruszecki, Jonathan Bell, Arjun Guha
https://arxiv.org/abs/2509.21891

AgentPack: A Dataset of Code Changes, Co-Authored by Agents and Humans
Fine-tuning large language models for code editing has typically relied on mining commits and pull requests. The working hypothesis has been that commit messages describe human intent in natural language, and patches to code describe the changes that implement that intent. However, much of the previously collected data is noisy: commit messages are terse, human-written commits commingle several unrelated edits, and many commits come from simple, rule-based bots. The recent adoption of softwar…

Human-like Navigation in a World Built for Humans
Bhargav Chandaka, Gloria X. Wang, Haozhe Chen, Henry Che, Albert J. Zhai, Shenlong Wang
https://arxiv.org/abs/2509.21189 https:…

Human-like Navigation in a World Built for Humans
When navigating in a man-made environment they haven't visited before--like an office building--humans employ behaviors such as reading signs and asking others for directions. These behaviors help humans reach their destinations efficiently by reducing the need to search through large areas. Existing robot navigation systems lack the ability to execute such behaviors and are thus highly inefficient at navigating within large environments. We present ReasonNav, a modular navigation system which …

💯 Scientists produce quantum entanglement-like results without entangled particles in new experiment
https://phys.org/news/2025-08-scientists-quantum-entanglement-results-entangled.html

Scientists produce quantum entanglement-like results without entangled particles in new experiment
In the everyday world that humans experience, objects behave in a predictable way, explained by classical physics. One of the important aspects of classical physics is that nothing, not even information, can travel faster than the speed of light. However, in the 1930s, scientists discovered that very small particles abide by some very different rules. One of the most mind-boggling behaviors exhibited by these particles is quantum entanglement—which Albert Einstein famously called "spooky acti…

Humans Perceive Wrong Narratives from AI Reasoning Texts
Mosh Levy, Zohar Elyoseph, Yoav Goldberg
https://arxiv.org/abs/2508.16599 https://arxiv.org/pdf/25…

Humans Perceive Wrong Narratives from AI Reasoning Texts
A new generation of AI models generates step-by-step reasoning text before producing an answer. This text appears to offer a human-readable window into their computation process, and is increasingly relied upon for transparency and interpretability. However, it is unclear whether human understanding of this text matches the model's actual computational process. In this paper, we investigate a necessary condition for correspondence: the ability of humans to identify which steps in a reasoning te…

Social World Models
Xuhui Zhou, Jiarui Liu, Akhila Yerukola, Hyunwoo Kim, Maarten Sap
https://arxiv.org/abs/2509.00559 https://arxiv.org/pdf/2509.00559

Social World Models
Humans intuitively navigate social interactions by simulating unspoken dynamics and reasoning about others' perspectives, even with limited information. In contrast, AI systems struggle to automatically structure and reason about these implicit social contexts. In this paper, we introduce a novel structured social world representation formalism (S3AP), designed to help AI systems reason more effectively about social dynamics. Following a POMDP-driven design, S3AP represents social interactions …

OpenAI says its reasoning system solved all 12 problems at the 2025 ICPC World Finals, with GPT-5 solving 11 and an experimental model solving the last (Maximilian Schreiner/The Decoder)
https://the-decoder.com/openai-outperforms…

OpenAI outperforms humans and Google at the world's top collegiate programming contest
An OpenAI system has solved every problem at the world’s most prestigious collegiate programming championship, outperforming not only human competitors but also Google's Deepmind model.

Assessing Human Cooperation for Enhancing Social Robot Navigation
Hariharan Arunachalam, Phani Teja Singamaneni, Rachid Alami
https://arxiv.org/abs/2508.21455 https://

Assessing Human Cooperation for Enhancing Social Robot Navigation
Socially aware robot navigation is a planning paradigm where the robot navigates in human environments and tries to adhere to social constraints while interacting with the humans in the scene. These navigation strategies were further improved using human prediction models, where the robot takes the potential future trajectory of humans while computing its own. Though these strategies significantly improve the robot's behavior, it faces difficulties from time to time when the human behaves in an…

Wrong Face, Wrong Move: The Social Dynamics of Emotion Misperception in Agent-Based Models
David Freire-Obreg\'on
https://arxiv.org/abs/2509.00080 https://

Wrong Face, Wrong Move: The Social Dynamics of Emotion Misperception in Agent-Based Models
The ability of humans to detect and respond to others' emotions is fundamental to understanding social behavior. Here, agents are instantiated with emotion classifiers of varying accuracy to study the impact of perceptual accuracy on emergent emotional and spatial behavior. Agents are visually represented with face photos from the KDEF database and endowed with one of three classifiers trained on the JAFFE (poor), CK+ (medium), or KDEF (high) datasets. Agents communicate locally on a 2D toroida…

Uncovering the Computational Ingredients of Human-Like Representations in LLMs
Zach Studdiford, Timothy T. Rogers, Kushin Mukherjee, Siddharth Suresh
https://arxiv.org/abs/2510.01030

Uncovering the Computational Ingredients of Human-Like Representations in LLMs
The ability to translate diverse patterns of inputs into structured patterns of behavior has been thought to rest on both humans' and machines' ability to learn robust representations of relevant concepts. The rapid advancement of transformer-based large language models (LLMs) has led to a diversity of computational ingredients -- architectures, fine tuning methods, and training datasets among others -- but it remains unclear which of these ingredients are most crucial for building models that …

Multi-Agent Data Visualization and Narrative Generation
Anton Wolter, Georgios Vidalakis, Michael Yu, Ankit Grover, Vaishali Dhanoa
https://arxiv.org/abs/2509.00481 https://

Multi-Agent Data Visualization and Narrative Generation
Recent advancements in the field of AI agents have impacted the way we work, enabling greater automation and collaboration between humans and agents. In the data visualization field, multi-agent systems can be useful for employing agents throughout the entire data-to-communication pipeline. We present a lightweight multi-agent system that automates the data analysis workflow, from data exploration to generating coherent visual narratives for insight communication. Our approach combines a hybrid…