Environments are the bottleneck for self-improving agents. Current terminal benchmarks were built for evaluation, not training; reinforcement learning requires a scalable pipeline, not just a dataset. We introduce Endless Terminals, a fully autonomous pipeline that procedurally generates terminal-use tasks without human annotation. The pipeline has four stages: generating diverse task descriptions, building and validating containerized environments, producing completion tests, and filtering for solvability. From this pipeline we obtain 3255 tasks spanning file operations, log management, data processing, scripting, and database operations. We train agents using vanilla PPO with binary episode level rewards and a minimal interaction loop: no retrieval, multi-agent coordination, or specialized tools. Despite this simplicity, models trained on Endless Terminals show substantial gains: on our held-out dev set, Llama-3.2-3B improves from 4.0% to 18.2%, Qwen2.5-7B from 10.7% to 53.3%, and Qwen3-8B-openthinker-sft from 42.6% to 59.0%. These improvements transfer to human-curated benchmarks: models trained on Endless Terminals show substantial gains on held out human curated benchmarks: on
Producing long, coherent video sequences with stable 3D structure remains a major challenge, particularly in streaming scenarios. Motivated by this, we introduce Endless World, a real-time framework for infinite, 3D-consistent video generation.To support infinite video generation, we introduce a conditional autoregressive training strategy that aligns newly generated content with existing video frames. This design preserves long-range dependencies while remaining computationally efficient, enabling real-time inference on a single GPU without additional training overhead.Moreover, our Endless World integrates global 3D-aware attention to provide continuous geometric guidance across time. Our 3D injection mechanism enforces physical plausibility and geometric consistency throughout extended sequences, addressing key challenges in long-horizon and dynamic scene synthesis.Extensive experiments demonstrate that Endless World produces long, stable, and visually coherent videos, achieving competitive or superior performance to existing methods in both visual fidelity and spatial consistency. Our project has been available on https://bwgzk-keke.github.io/EndlessWorld/.
Procedural Content Generation (PCG) enables game content to be created algorithmically without direct manual level-design effort, but it introduces a serious evaluation problem: generated content may become unbalanced, blocked, repetitive, or technically unsolvable. This paper presents Momentum, an endless-runner game that integrates runtime terrain generation, environment object spawning, and autonomous agent-based evaluation into a single gameplay loop. Ground tiles and environmental objects are generated dynamically as the player advances, object placement follows a constraint-driven mechanism inspired by Wave Function Collapse (WFC), and the runtime navigation surface is rebuilt asynchronously to remain consistent with the streamed environment. Two autonomous evaluation agents move ahead of the player and inspect the generated path: an aerial scanner that examines the corridor geometrically, and a ground-traversal agent that validates the same region from a navigational perspective. The evaluation pipeline combines ray casting, volumetric physics sweeps, obstacle-layer filtering, and structured crash reporting to identify problematic generated scenarios before they reach the pl
Large language models (LLMs) are increasingly used to support software development, but their practical usefulness in applied game-development settings remains underexplored, especially when generated code must be integrated into an existing game software system. This paper presents an exploratory empirical case study of GPT-4o in a custom Python/Pygame endless runner. The study examines six selected development tasks: three localized refactoring tasks and three tasks involving gameplay feature generation. The resulting implementations were evaluated using software metrics, unit tests, and manual gameplay assessments. In this case study, all three selected refactoring tasks were completed successfully in functional terms, whereas only one of the three selected gameplay feature generation tasks resulted in a correctly integrated feature. The findings suggest that, in this setting, GPT-4o handled localized transformations more reliably than tasks requiring new gameplay interactions across multiple existing systems. Given the exploratory single-case design, these results are best interpreted as indicative observations rather than as generalizable evidence of category-level model perfo
The grand vision of enabling persistent, large-scale 3D visual geometry understanding is shackled by the irreconcilable demands of scalability and long-term stability. While offline models like VGGT achieve inspiring geometry capability, their batch-based nature renders them irrelevant for live systems. Streaming architectures, though the intended solution for live operation, have proven inadequate. Existing methods either fail to support truly infinite-horizon inputs or suffer from catastrophic drift over long sequences. We shatter this long-standing dilemma with InfiniteVGGT, a causal visual geometry transformer that operationalizes the concept of a rolling memory through a bounded yet adaptive and perpetually expressive KV cache. Capitalizing on this, we devise a training-free, attention-agnostic pruning strategy that intelligently discards obsolete information, effectively ``rolling'' the memory forward with each new frame. Fully compatible with FlashAttention, InfiniteVGGT finally alleviates the compromise, enabling infinite-horizon streaming while outperforming existing streaming methods in long-term stability. The ultimate test for such a system is its performance over a tru
The Endless Tuning is a design method for a reliable deployment of artificial intelligence based on a double mirroring process, which pursues both the goals of avoiding human replacement and filling the so-called responsibility gap (Matthias 2004). Originally depicted in (Fabris et al. 2024) and ensuing the relational approach urged therein, it was then actualized in a protocol, implemented in three prototypical applications regarding decision-making processes (respectively: loan granting, pneumonia diagnosis, and art style recognition) and tested with such as many domain experts. Step by step illustrating the protocol, giving insights concretely showing a different voice (Gilligan 1993) in the ethics of artificial intelligence, a philosophical account of technical choices (e.g., a reversed and hermeneutic deployment of XAI algorithms) will be provided in the present study together with the results of the experiments, focusing on user experience rather than statistical accuracy. Even thoroughly employing deep learning models, full control was perceived by the interviewees in the decision-making setting, while it appeared that a bridge can be built between accountability and liabili
Kagome-lattice crystal is crucial in quantum materials research, exhibiting unique transport properties due to its rich band structure and the presence of nodal lines and rings. Here, we investigate the electronic transport properties and perform first-principles calculations for Ni$_{3}$In$_{2}$Se$_{2}$ kagome topological semimetal. First-principle calculations indicate six endless Dirac nodal lines and two nodal rings with a $π$-Berry phase in the Ni$_{3}$In$_{2}$Se$_{2}$ compound. The temperature-dependent resistivity is dominated by two scattering mechanisms: $s$-$d$ interband scattering occurs below 50 K, while electron-phonon ($e$-$p$) scattering is observed above 50 K. The magnetoresistance (MR) curve aligns with the theory of extended Kohler's rule, suggesting multiple scattering origins and temperature-dependent carrier densities. A maximum MR of 120\% at 2 K and 9 T, with a maximum estimated mobility of approximately 3000 cm$^{2}$V$^{-1}$s$^{-1}$ are observed. The Ni atom's hole-like d$_{x^{2}-y^{2} }$ and electron-like d$_{z^{2}}$ orbitals exhibit peaks and valleys, forming a local indirect-type band gap near the Fermi level (E$_{F}$). This configuration enhances the mot
Panoramic Image Generation (PIG) aims to create coherent images of arbitrary lengths. Most existing methods fall in the joint diffusion paradigm, but their complex and heuristic crop connection designs often limit their ability to achieve multilevel coherence. By deconstructing this challenge into its core components, we find it naturally aligns with next-token prediction, leading us to adopt an autoregressive (AR) paradigm for PIG modeling. However, existing visual AR (VAR) models are limited to fixed-size generation, lacking the capability to produce panoramic images. In this paper, we propose PanoLlama, a novel framework that achieves endless and coherent panorama generation with the autoregressive paradigm. Our approach develops a training-free strategy that utilizes token redirection to overcome the size limitations of existing VAR models, enabling next-crop prediction in both horizontal and vertical directions. This refreshes the PIG pipeline while achieving SOTA performance in coherence (47.50%), fidelity(28.16%), and aesthetics (15%). Additionally, PanoLlama supports applications other PIG methods cannot achieve, including mask-free layout control, multi-scale and multi-gui
Memory Gym presents a suite of 2D partially observable environments, namely Mortar Mayhem, Mystery Path, and Searing Spotlights, designed to benchmark memory capabilities in decision-making agents. These environments, originally with finite tasks, are expanded into innovative, endless formats, mirroring the escalating challenges of cumulative memory games such as "I packed my bag". This progression in task design shifts the focus from merely assessing sample efficiency to also probing the levels of memory effectiveness in dynamic, prolonged scenarios. To address the gap in available memory-based Deep Reinforcement Learning baselines, we introduce an implementation within the open-source CleanRL library that integrates Transformer-XL (TrXL) with Proximal Policy Optimization. This approach utilizes TrXL as a form of episodic memory, employing a sliding window technique. Our comparative study between the Gated Recurrent Unit (GRU) and TrXL reveals varied performances across our finite and endless tasks. TrXL, on the finite environments, demonstrates superior effectiveness over GRU, but only when utilizing an auxiliary loss to reconstruct observations. Notably, GRU makes a remarkable r
Despite extensive safety measures, LLMs are vulnerable to adversarial inputs, or jailbreaks, which can elicit unsafe behaviors. In this work, we introduce bijection learning, a powerful attack algorithm which automatically fuzzes LLMs for safety vulnerabilities using randomly-generated encodings whose complexity can be tightly controlled. We leverage in-context learning to teach models bijective encodings, pass encoded queries to the model to bypass built-in safety mechanisms, and finally decode responses back into English. Our attack is extremely effective on a wide range of frontier language models. Moreover, by controlling complexity parameters such as number of key-value mappings in the encodings, we find a close relationship between the capability level of the attacked LLM and the average complexity of the most effective bijection attacks. Our work highlights that new vulnerabilities in frontier models can emerge with scale: more capable models are more severely jailbroken by bijection attacks.
In this paper, we revisit endless online level generation with the recently proposed experience-driven procedural content generation via reinforcement learning (EDRL) framework. Inspired by an observation that EDRL tends to generate recurrent patterns, we formulate a notion of state space closure which makes any stochastic state appeared possibly in an infinite-horizon online generation process can be found within a finite-horizon. Through theoretical analysis, we find that even though state space closure arises a concern about diversity, it generalises EDRL trained with a finite-horizon to the infinite-horizon scenario without deterioration of content quality. Moreover, we verify the quality and the diversity of contents generated by EDRL via empirical studies, on the widely used Super Mario Bros. benchmark. Experimental results reveal that the diversity of levels generated by EDRL is limited due to the state space closure, whereas their quality does not deteriorate in a horizon which is longer than the one specified in the training. Concluding our outcomes and analysis, future work on endless online level generation via reinforcement learning should address the issue of diversity
Topological semimetals are a frontier of quantum materials. In multi-band electronic systems, topological band-crossings can form closed curves, known as nodal lines. In the presence of spin-orbit coupling and/or symmetry-breaking operations, topological nodal lines can break into Dirac/Weyl nodes and give rise to novel transport properties, such as the chiral anomaly and giant anomalous Hall effect. Recently the time-reversal symmetry-breaking induced Weyl fermions are observed in a kagome-metal Co3Sn2S2, triggering interests in nodal-line excitations in multiband kagome systems. Here, using first-principles calculations and symmetry based indicator theories, we find six endless nodal lines along the stacking direction of kagome layers and two nodal rings in the kagome plane in nonmagnetic Ni3 In2 S2 . The linear dipsersive electronic structure, confirmed by angle-resolved photoemission spectroscopy, induces large magnetoresistance up to 2000% at 9 T. Our results establish a diverse topological landscape of multi-band kagome metals.
We propose a simple deterministic dynamic equation and reveal the mechanism of large-scale endless evolvement of spatial density inhomogeneity in active nematic. We determine the phase regions analytically. The interplay of density, magnitude of nematic order, and nematic director is crucial for the long-wave-length instability and the emergence of seemingly fluctuated collective motions. Ordered nematic domains can absorb particles, grow and divide endlessly. The present finding extends our understanding of the large-scale and seemingly fluctuated organization in active fluids.
Scientific research in the United States could receive a large increase in federal funding--up to 100 billion dollars over five years -- if proposed legislation entitled the Endless Frontiers Act becomes law. This bipartisan and bicameral bill, introduced in May 2020 by Senators Chuck Schumer (D-NY) and Todd Young (R-IN) and Congressmen Ro Khanna (D-CA) and Mike Gallagher (R-WI), is intended to expand the funding of the physical sciences, engineering, and technology at the National Science Foundation (NSF) and create a new Technology Directorate focused on use-inspired research. In addition to provisions to protect the NSF's current missions, a minimum of 15\% of the newly appropriated funds would be used to enhance NSF's basic science portfolio. The Endless Frontier Act offers a rare opportunity to enhance the breadth and financial support of the American research enterprise. In this essay, we consider the benefits and the liabilities of the proposed legislation and recommend changes that would further strengthen it.
We provide a rigorous analysis for the so-called endlessly continuable germs of holomorphic functions or in other words, the Ecalle's resurgent functions. We follow and complete an approach due to Pham, based on the notion of discrete filtered set and the associated Riemann surface defined as the space of convenient homotopy classes of paths. Our main contribution consists in a complete though simple proof of the stability under convolution product of the space of endlessly continuable germs.
We introduce a self-avoiding walk model for which end-effects are completely eliminated. We enumerate the number of these walks for various lattices in dimensions two and three, and use these enumerations to study the properties of this model. We find that endless self-avoiding walks have the same connective constant as self-avoiding walks, and the same Flory exponent $ν$. However, there is no power law correction to the exponential number growth for this new model, i.e. the critical exponent $γ= 1$ exactly. In addition, we have convincing numerical evidence to support the hypothesis that the amplitude for the number growth is a universal quantity, and somewhat weaker evidence which suggests that the number growth has no analytic corrections to scaling. The technique by which end-effects are eliminated may be generalised to other models of polymers such as interacting self-avoiding walks.
Consider the set of isoenergetically degenerate integrable Hamiltonians with two degrees of freedom. We show that a cusp-generic perturbation of a generic Hamiltonian in this set gives rise to meandering invariant tori - embedded Lagrangian tori which are not graphs. Moreover, an exponentially dense subset of perturbations admits higher order meandering tori, of all orders from two to infinity. These infinite order meanders have an endless nested structure.
This is a research exploring existing models and fine tuning them to combine a YOLOv8 segmentation model, a LSTM model trained on hand point motion sequence and a ASR (whisper-base) to extract enough data for a LLM (TinyLLaMa) to predict the recipe and generate text creating a step by step guide for the cooking procedure. All the data were gathered by the author for a robust task specific system to perform best in complex and challenging environments proving the extension and endless application of computer vision in daily activities such as kitchen work. This work extends the field for many more crucial task of our day to day life.
Private Everlasting Prediction (PEP), recently introduced by Naor et al. [2023], is a model for differentially private learning in which the learner never publicly releases a hypothesis. Instead, it provides black-box access to a "prediction oracle" that can predict the labels of an endless stream of unlabeled examples drawn from the underlying distribution. Importantly, PEP provides privacy both for the initial training set and for the endless stream of classification queries. We present two conceptual modifications to the definition of PEP, as well as new constructions exhibiting significant improvements over prior work. Specifically, (1) Robustness: PEP only guarantees accuracy provided that all the classification queries are drawn from the correct underlying distribution. A few out-of-distribution queries might break the validity of the prediction oracle for future queries, even for future queries which are sampled from the correct distribution. We incorporate robustness against such poisoning attacks into the definition of PEP, and show how to obtain it. (2) Dependence of the privacy parameter $δ$ in the time horizon: We present a relaxed privacy definition, suitable for PEP,
We prove that formal WKB solutions of Schrödinger equations on Riemann surfaces are resurgent. Specifically, they are Borel summable in almost all directions and their Borel transforms admit endless analytic continuation away from a discrete subset of singularities. Our approach is purely geometric, relying on understanding the global geometry of complex flows of meromorphic vector fields using techniques from holomorphic Lie groupoids and the geometry of spectral curves. This framework provides a fully geometric description of the Borel plane, Borel singularities, and the Stokes rays. In doing so, we introduce a geometric perspective on resurgence theory.