Andrej Karpathy built a simple automation pipeline for AI agents to optimize training in 5-minute increments.

Scientific publication compresses a branching, iterative research process into a linear narrative, discarding the majority of what was discovered along the way.

To guide evolution, we derive 'textual gradients,' structured natural language feedback from execution traces, to pinpoint failures and suggest granular modifications.

The duo had jump-started the AI-for-Erdős craze late last year by prompting a free version of ChatGPT with open problems chosen at random from the Erdős problems website.

An AI researcher subsequently gifted them each a ChatGPT Pro subscription to encourage their 'vibe mathing.'

An AI researcher subsequently gifted them each a ChatGPT Pro subscription to encourage their 'vibe mathing.'

_Self-reported score with custom Anthropic scaffold._ SWEPro were evaluated with the mini-swe-agent scaffold. However, we use the scores reported by Anthropic for Opus with the max thinking efforts due to frequent timeouts during our evaluation trials.

In our internal evals and testing, medium effort achieved slightly lower intelligence with significantly less latency for the majority of tasks.

We examine whether AI capabilities are accelerating by fitting statistical models to benchmark performance over time, and comparing their predictive accuracies.

We have been calling this the 'reasoning' / 'non-reasoning' split, but this is not a perfectly clean dichotomy. Several correlated but not strictly identical changes happened over the same few months: scaling inference compute, heavier use of RL in post-training, and models producing reasoning tokens.

We pre-selected the 6-month horizon as our primary metric, balancing genuine forecasting distance against the limited date range of our data.

We study a mix of Hugging Face downloads and model derivatives, inference market share, performance metrics and more to make a comprehensive picture of the ecosystem.

The H100-equivalent unit uses a chip's highest 8-bit operation/second specifications to convert between chips. The actual utility of a particular chip depend on workload assumptions, so H100e does not perfectly reflect real-world performance differences across chip types.

The standard autoresearch loop (brainstorm from code, run experiments, check metrics) works when the optimization surface is visible in the source. The Liquid results prove that. But for problems where the codebase doesn't contain enough information to generate good hypotheses, giving the agent access to papers and competing implementations changes what it tries.

Gemini 3 Flash achieves the highest score of 24.0%

As AI moves from a destination to a feature, our methodology will need to shift.

two participants gave it 9/10 and one "11/10"

Large language models (LLMs) sometimes appear to exhibit emotional reactions. We investigate why this is the case in Claude Sonnet 4.5 and explore implications for alignment-relevant behavior.

We convert chip computing capabilities into H100 equivalents (H100e) based on their relative FLOP/s specifications, specifically their maximum 8-bit specification.

leading baselines achieve only about half the accuracy at the same efficiency

To enable true process-level verification, we audit fine-grained intermediate states rather than just final answers, and quantify efficiency via an overthinking metric relative to human trajectories.

We introduce Iterative Reward Calibration, a methodology for designing per-turn rewards using empirical discriminative analysis of rollout data

We introduce Iterative Reward Calibration, a methodology for designing per-turn rewards using empirical discriminative analysis of rollout data

We recruited 12 active researchers by word of mouth followed bysnowball sampling [ 53 ]

Interviews were video and audio recorded. We transcribed the audio using OpenAI's Whisper automatic speech recognition system and anonymized the transcript before analysis.

Immediately after the focused reading task, we conducted a short interview asking participants to reflect on their experience with both tasks (Appendix J.2), followed by a post-study survey (Appendix J.3).

The study concluded with a 15-minute semi-structured interview. During the interview, participants saw screenshots from the three conditions and were asked which they preferred and disliked, why, what they wished the interface had, what influenced their skimming, and how they normally skimmed texts.

After the interviews, we analyzed the data using the process described in Appendix B

We used these mock-ups as design probes [31] to inspire ideation and elicit creative responses. Specifically, we asked participants to compare and contrast alternative mock-ups and reflect on how they could be used or improved to support their known or emerging synthesis and information-foraging goals.

In the second part of the session, we provided participants with mock-ups of possible reifications of cross-document relationships that might help them synthesize information across abstracts.

In the first part of the session, we asked participants about their strategies for selecting publication venues for their manuscript submissions, how they identify and synthesize information from venues, their approaches to writing manuscripts, and finally, the technology they have used to help with these processes, current technology shortcomings, and ideas for addressing these challenges.

Sessions, which were held on Zoom, lasted 55 minutes on average. Participants were compensated with $15 USD.

The interview sessions were divided into two parts: an open-ended semi-structured interview about their backgrounds and practices, followed by feedback on a range of mock-ups, including novel reified relationships between analogous sentences in different abstracts (Figure 2).

In order to determine (1) the context in which we might offer novel views of scientific abstracts and (2) the intelligibility of various novel prototype designs for reifying cross-abstract relationships, we conducted a formative interview study with 12 active researchers (see Appendix A for participant information).

or example, Lunde-berg, Fox, and Puncochar [1994] argue that the reason why somestudies do not find gender differences in confidence on generalknowledge is because it is not in the masculine domain

Using discourse analysis, the article identifies “interpretative repertoires” (Gil-bert and Mulkay 1984) and linguistic resources that are employed by the authors of profes-sional journal articles and blogs and that characterize makerspaces in particular ways. In atheory of discourse, librarians who identify themselves within these discursive constructsbecome subjects of those discourses, thus reproducing particular ways of thinking aboutmakerspaces

educationaldesign research methodology.

. We note, however, a few limitations ofour work

the sample differed along a number of other dimensions (e.g., non-DDP minestended to produce gold; DDP mines produce 3T)

IPIS research teams collect quantitative and qualitative data through a combination of observations andinterviews with a selection of stakeholders at and around ASM sites, support villages, and trading hubs.Data collection methods and verification rely extensively on triangulation of sources, and intervieweesinclude artisanal miners, representatives of cooperatives, heads of miner camps, state agents and CSOrepresentatives.

IPIS visited 354 active mining sites, which employ an estimated 55,604 miners (Table A). Gold mines ac-counted for 86% of mines visited, followed by tin, tantalum, and tungsten (3T) mines, which accountedfor 20% of mines (Table B). These proportions are consistent with the larger IPIS database on ASM sitesin eastern DRC

When comparing data collected on the mining sites during field visits with data from mobilesurveys, it is important to keep in mind the two levels of analysis employed: mining sites and individualexperience and perception.It worth noting that the low number of gold mining sites covered by due diligence programmes meansthat comparing DDP and non-DDP mining sites will inherently be subject to any difference between 3Tand gold exploitations.

IPIS surveyors visited mine sites andtrading centres (so-called points de vente) in two consecutive field missions, during which they collectedphone numbers of community members, made observations, and conducted interviews with several keyinformants so as to complete an extensive questionnaire on each mine site (using the OpenDataKit tool)

y combining social science methods and underwater reef surveys we identify a number of countervailing challenges and opportunities presented by globalization that both nurture and suppress the island's resilience to high amplitude, low-frequency disturbances like tsunamis

Investigating social structures through the use of network or graphs Networked structures Usually called nodes ((individual actors, people, or things within the network) Connections between nodes: Edges or Links Focus on relationships between actors in addition to the attributes of actors Extensively used in mapping out social networks (Twitter, Facebook) Examples: Palantir, Analyst Notebook, MISP and Maltego

two-tailed t-test

Significance levels

between-subject

within

voxel

T2*-sensitive functional imag-ing

3-T Siemens Trio MRI scan-ner

fixation cross

inter stimulus intervals

T1-weighted images

cerebellum

cortex

The waterfall methodology for software development is rapidly losing popularity, while the Agile methodology is increasingly used for software development by companies around the world.

Users are shown four animated GIFs, each correspondingto modifying an attribute for a given image.

ee Sec. 4.2.2 for user studiesthat further demonstrate this

AttFind proceeds as follows: At each iteration it consid-ers all K style coordinates and calculates their effect on theprobability of y.3 4 It then selects the coordinate with largesteffect, and removes all images where changing this coordi-nate had a large effect on their probability to belong to classy

AttFind takes as input the trained model and a set of N im-ages whose predicted label by C is different from y. Foreach class y (e.g., y=“cat” or y=“dog”), AttFind then findsa set Sy of M style coordinates (i.e., Sy ⊂ [1, . . . , K] and|Sy|= M ), such that changing these coordinates increasesthe average probability of the class y on these images.

(the VTA is also part ofthis system, but is too small to image with standard fMRImethods, but see [35] for successful imaging methods).

How to Conduct Agile Market Research for Your Digital Product

rank correlation analysis, using Kendall's tau-c. For this purpose the three urban renewal status classes are assumed to constitute a scale. Evidence that such an assumption is reasonable is present in Tables 1 and 3.

Two other controls having to do with the socioeconomic level of the resident pop- ulation are used

Since the significance of the number of functions varies with the number of all other functions (i.e., the size of the employed labor force), it should be ex- pressed as a ratio to the latter. Hence the lower the ratio of managers, proprietors, and officials7 to the employed labor force the greater is the concentration of power. (This measure will hereafter be called the MPO ratio.

acknowledge the value of the knowledge thatis held by research communities in the data collection process

requested consent to quote brief,anonymized portions of conversations to provide anecdotalcontext.

In this practice, only errors from outside the program's control are to be handled (such as user input); the software itself, as well as data from within the program's line of defense, are to be trusted in this methodology.

n any nonviolent campaign there are four basic steps: collection of the facts to determine whether injustices exist; negotiation; self purification; and direct action. We have gone through all these steps in Birmingham

methodological arms race,

Analyses linking Pentecostalism to the pursuit of personal gain through the breakdown of social relationships are problematic on at least two fronts. First, the vast majority of these studies are based largely, if not exclusively, on sermons and interviews with church leaders. While these are certainly an important part of any anthropological discussion of Pentecostal ritual life, in the absence of a robust ethnographic engagement with those who spend their time listening to these messages, it is impossible to determine the extent to which believers are putting their leaders' words into practice (see Engelke 2010).

“To me, this is symptomatic of a much larger problem of transparency within the company. Nobody is forthcoming with information that dramatically affects editorial,” Binkowski said. “One of those things was me not knowing if I was in trouble.”

When I asked Barbara to comment on the GoFundMe page, she noticed her erasure. “Was surprised to see my life’s work described as having been ‘a small one-­person effort,’”

When I asked how many articles she’d written for the site, she came back with a “verified count” of 1,905. She told me how she came to that number: “By examining every Snopes.com HTML file on my computer, rereading every email David and I exchanged from 1997 until now, and in cases where doubt still existed, examining my research files. The task took a week, but I am satisfied I now have a fair list and that all lurking doubles (a result of David’s penchant for renaming files) have been excised.”

Methodology The classic OSINT methodology you will find everywhere is strait-forward: Define requirements: What are you looking for? Retrieve data Analyze the information gathered Pivoting & Reporting: Either define new requirements by pivoting on data just gathered or end the investigation and write the report.

Appendix A. The survey instrument

SURVEY DATA ON NUCLEAR POWER

To be sure, the topicality, novelty or potential benefits of a given line of research might help it attract notice and support, butscientific researchfundamentally stands or falls on the thoroughness with which activities and reasoning can be tied together. You just can’t get in the game without a solid methodology.

telephone interviews with 37 participants

Because of both the content that people upload and the behavioral traces that they leavebehind, social network sites have unprecedented quantities of data concerning humaninteraction. This presents unique opportunities and challenges. On one hand, SNSs offera vibrant “living lab” and access to behavioral data at a scale inconceivable to manysocial scientists. On the other, the data that are available present serious research ethicsquestions and introduce new types of biases that must be examined (boyd and Crawford2012)

For those of us who believe that social network sites are socio-technical systems, in whichsocial and technical factors shape one another, failing to describe the site under studyignores the fact that the technological constraints and affordances of a site will shapeuser practices and that social norms will emerge over time. Not including informationabout what the feature set was at the time of data collection forecloses the possibility ofidentifying patterns that emerge over time and through the accumulated scholarshipacross a range of sites and user samples. Unfortunately, because they have no knowledgeabout how things will continue to evolve and which features will becomeimportant to track, researchers may not be able to identify the salient features to reportand may struggle with devoting scarce publication space to these details, but this doesn’tundermine the importance of conscientious consideration towards describing the artifactbeing analyzed.

In order to produce scholarship that will be enduring, the onus is on social mediaresearchers to describe the technological artifact that they are analyzing with as muchcare as survey researchers take in describing the population sampled, and with as muchdetail as ethnographers use when describing their field site. This is not to say thatresearchers must continue to describe technologies as if no one knows what they are—weare beyond the point where researchers must explain how electronic mail or “email” islike or unlike postal mail. But, rather, researchers must clearly describe the socio-technical context of the particular site, service, or application their scholarship isaddressing. In addition to attending to the technology itself, and the interchange betweentechnical and social processes, we believe SNS researchers should make a concertedeffort to include the date of data collection and to describe the site at the moment of datacollection and the relevant practices of its users. These descriptions will enable laterresearchers to synthesize across studies to identify patterns, much in the same wayreporting exact effect sizes allows for future meta-analyses

I think of research as a conversation, and it really is very much like a conversation. No single person dominates it, but what does happen is when you interject something, when you contribute something to a conversation, you want to be understood, you want to be heard, you would like people to pay attention, you would like it to have some influence on the way the conversation goes. You don't control it. But thinking in conversational terms and trying to say something that is interesting as a criteria, not merely publishable but actually is interesting -- that's been part of what moved me.

studied the social network structure underlying Youtube videos

Emma’s multiplicity of subplots, andits preoccupation with the reading, rereading and misreading of writing within,and events internal to, the text, renders the novel a manifesto for Austen’sapproach to the‘‘judicious’’, critical reading necessary to understanding thefunction of literary influence in her fiction

The catalyst forthe novel, however, seems to have been a straightforward reaction to a newwork by an author Austen considered her competition*the Scottish MaryBrunton’sDiscipline(1814).Disciplineis a fictional autobiography with the strong religious themes ofsin, repentance and redemption.

Still, it may not be others' actual behavior that drives ourown ad-dictive behavior, but our perceptions of their behavior, where the twoconflict.

This suggested that for this survivor, those feelings were the most powerful aspects of her experience.  

snowball sampling method where we initially asked 17 re-spondents about their social and expert networks. This in-volved asking informants to name the seven most importantpersons in their lives, starting with the most important, outsideoftheir household

Austen allows Emma to imaginatively misattribute herself. In doing so,she offers the reader a literary red herring. While Harriet may fall in and out oflove as if she is subject to one of Puck’s spells,Emmatakes its cues from adifferent Shakespearean comedy.24Emma, who has‘‘very little intention of ever marrying at all’’, yet is happyto consider Frank Churchill as a potential husband (84), resembles Olivia, the‘‘too proud’’heiress of Shakespeare’sTwelfth Night, whose resolution to live‘‘like a cloistress’’is quickly abandoned when she meets Viola, disguised as aboy.25

It is not to be expected that any character withinEmmamight be able toexercise the kind of judgment of its creator or perform the kind of judiciousreading that Austen’s text ultimately demands. This does not prevent Austenfrom demonstrating how her characters can betaughtto read and to judgeclearly.

If we enlarge our understanding of the concept of‘‘influence’’, we canbegin to see the ways in which artistically unremarkable, canonicallydisregarded works inform the development even of masterpieces. Ros Ballastercorrectly states that:[...] most women novelists of the eighteenth century tended to locatetheir own writing in relation to a strong line of male predecessors orcontemporaries [...] if women read each other’s work they did not, forthe most part, openly acknowledge influence.16Jane Austen is the exception to this rule. Far from shamefacedly concealing herdebt to Brunton’s novel, on the contrary, Austen’s linguistic allusions toDisciplineinEmmadraw the reader’s attention to the two novels’intimateconnection

Such active, criticalreading, of course, distinguishes the work still expected of students andscholars of literature, and Austen’s assumption of this ability in her readers wasunderpinned by historic changes in the study of literature.

‘‘I do not write for such dull Elves’’, wrote Jane Austen to her sisterCassandra in 1813,‘‘As have not a great deal of Ingenuity themselves’’.

look at the edges, the connections between the nodes.

descriptive analysis