Skip to content
Log in
Sponsored Design Multi-Agent AI Systems Using MCP and A2A: Engineer your own Pyt Design Multi-Agent AI Systems Using MCP and A2A: Engineer your own Pyt

Category

small language model

3,573 papers

#small language model Open access Sep 2026

LLM Parkinsonism: Executive-Control Failure, Token-Inefficient Persistence, and a Global Executive Control Architecture for Autonomous Language-Model Agents

Large language models (LLMs) can plan, write code, call tools, manipulate software, and execute long-horizon workflows. Yet strong local competence does not guarantee project-level executive control. An agent can make large, useful early advances and then shift toward progressively smaller, lower-value actions—addition...

Dongsheng Xiao · 0 citations
#explainable ai Open access Sep 2026

PREreview of "The widening evaluation gap in medical large language model research 2023 to 2026"

This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/22734670. What the paper claims This is an evidence map of 11,628 PubMed-indexed records on generative language models in healthcare, January 2023 to June 2026, built to answer one...

Evgeny V. Arsentyev · 0 citations
#small language model Dataset Open access Sep 2026

AccessDetector: Android accessibility audit dataset (720 screens x 3 VLMs, 9-tag taxonomy, human-validated ground truth)

Dataset and analytics produced by the AccessDetector pipeline: 835 captured Android screens across 138 apps with valid captures (from 140 explored), audited by 3 vision-language models (GPT-5.4, Claude Sonnet 4.6, Gemini 2.5 Pro) under a 9-tag accessibility taxonomy. The analytic universe of the paper is the 720-screen...

Benjamin Gonzalez Briones · 0 citations
#small language model Open access Sep 2026

What Lifts the Liquid Is Not Surface Tension but Its Difference ── Wine at 12 per cent is 0.039445 N/m; evaporation to 10 raises it to 0.042828 ── the difference of 0.003383 N/m lifts a 20 μm film by 17.25 mm ── pure water has the largest tension and no difference, so it makes no tears ── [Paper 452]

The greater the surface tension, the higher the liquid climbs──so it is thought. This paper shows that what lifts the liquid is not the tension but its difference──which is why pure water, with the largest tension of all, makes no tears. No new theorem or law is claimed. Scope of this paper (scope note): No new theorem...

Yuuki Yamagishi · 0 citations
#small language model Dataset Open access Sep 2026

An audit of presentation-form corruption in 341 public Arabic datasets

A measured sweep of 341 public Arabic-language datasets on the Hugging Face Hub for text corrupted by the widespread arabic_reshaper.reshape() + python-bidi get_display() pre-shaping recipe, which stores Arabic as Unicode presentation forms in visual order rather than as logical-order text. Method. Datasets were read t...

Syamjith NK · 0 citations
#small language model Open access Sep 2026

An Edge-AI-Driven Crystal Structure Exploration Tool for Secondary School Chemistry: Design and Implementation

Crystal structure is one of the most abstract topics in secondary school chemistry. Textbookspresent three-dimensional lattices as flat diagrams, and physical models cannot be modified, sostudents often memorize conclusions such as "NaCl has a coordination number of 6" withoutunderstanding how structure determines prop...

Kerong He · 0 citations
#small language model Open access Sep 2026

Both Improved 7.2446-Fold and the Ratio Never Moved ── Improving at two per cent a year, capability rises 7.2446-fold over a century while the ratio stays at exactly 1.000000000 ── let one side stop and the odds move from 0.5 to 0.878803 ── one runs not to overtake but not to fall ── [Paper 438]

Improving is expected to bring advantage. This paper shows that if the other improves as much, nothing changes──and that stopping nonetheless costs a great deal. No new theorem or law is claimed. Scope of this paper (scope note): No new theorem or law is claimed──the Red Queen hypothesis, coevolution and relative fitne...

Yuuki Yamagishi · 0 citations
#small language model Open access Sep 2026

Raising the Lift Coefficient by Sixty Per Cent Lowers the Stall Speed by Twenty ── Taking the maximum lift coefficient from 1.5 to 2.5, a factor of 1.6667, takes the stall speed only from 99.3694 to 76.9712 km/h, a factor of 0.7746 ── because it acts through a square root ── and 0.4518 at 25 degrees is worse than 1.2990 at 10 ── [Paper 458]

Fit a high-lift device and you fly correspondingly slower──so it is thought. This paper shows that the lift coefficient acts on the stall speed only through a square root──a factor of 1.6667 buys a drop of 22.5 per cent. No new theorem or law is claimed. Scope of this paper (scope note): No new theorem or law is claime...

Yuuki Yamagishi · 0 citations
#small language model Open access Sep 2026

The Same Proportional Increase Buys Less at the Top ── Raising the compression ratio by 1.3333 gains +0.0531 in efficiency starting from 6 but only +0.0359 starting from 16 ── the ratio is 1.480429, matching the closed form (r_2/r_1)^gamma-1 exactly ── an efficiency of 0.70 needs a ratio of 20.2860 ── [Paper 468]

Raising the compression ratio raises efficiency──so it is said. This paper shows that the same proportional increase buys 1.480429 times less at the top──and that the ratio has a closed form. No new theorem or law is claimed. Scope of this paper (scope note): No new theorem or law is claimed──the theoretical efficiency...

Yuuki Yamagishi · 0 citations
#small language model Open access Sep 2026

333 Primordial Constants: The Early Genetic Library of Huancaidie Civilizational Generative Syntax

This research monograph presents the 333 Primordial Constants Project as an interdisciplinary framework for studying the minimal constraints, organizing principles, and generative conditions that may underlie physical structure, mathematics, cognition, life, narrative, civilization, meaning, symbolic systems, agency, a...

政恩 馮 · 0 citations
#small language model Open access Sep 2026

What Sets the Speed Is the Power, Not the Gear ── 200 W balances at 32.13 km/h, far above the 13.46 km/h a 34x28 reaches ── what a gear selects is not speed but cadence ── reaching 25 km/h in 34x28 demands 167.2 rpm ── [Paper 449]

A low gear is slow and a high gear is fast──so it is thought. This paper shows that power sets the speed and the gear only selects the cadence──so “a lower gear makes you slower” is true only when the power drops. No new theorem or law is claimed. Scope of this paper (scope note): No new theorem or law is claimed──gear...

Yuuki Yamagishi · 0 citations
#small language model Open access Sep 2026

White Is Made by a Difference, Not by a Colour ── The same "white powders" scatter as 1 to 0.006944 once mixed into a resin, a factor of 144 ── wet the paper and the difference falls from 1.70 to 1.37, taking 35.1 per cent of the scattering with it ── match the index and the white vanishes exactly ── [Paper 442]

A white powder is taken to be white. This paper shows that whiteness is not a property of the powder──it is made by the difference from its surroundings, and vanishes when the difference does. No new theorem or law is claimed. Scope of this paper (scope note): No new theorem or law is claimed──light scattering, hiding...

Yuuki Yamagishi · 0 citations

From tech blogs

See all →

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.