Upload KnowledgeClaw batch (500 files)

.gitattributes

@@ -58,3 +58,142 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/f2445cd185172a10.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/8a2d9d32ad8a8c93.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/8c3457d4bc0fbf69.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/1aa610d1cf415b6e.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/7065c51a1036c2ec.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/6dde9e44672035d0.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/ad49b25686b3c8d0.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/3a67e714e5463107.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/c27673a72028df1e.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/8be38490fe66721d.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/1aba209b0e45c4a9.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/ce381baca9e0b98e.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/31590a701f344f8b.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/5aa01453fde49f59.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/0f97b0c4b5dc07a1.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/911a1a74161c1d55.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/c3b7d19a937ca974.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/e3ce822a587fab4c.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/f59f85797482721f.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/48d1dc28684ace55.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/ce25cac5c7346068.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/7fa78487dc4976c4.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/fdc0cae06eb3d7fa.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/75d1cea6b61b9628.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/7107c73f8afa3151.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/923fdbdaa97e792b.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/9d1efe0e5c61efb1.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/0bb78f6caf5c69f0.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/3971ed8835a193f9.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/2b58cad4ea8f4e7c.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/e646025167dcad42.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/ad90363d4f0110ec.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/62d47d53f5b7713d.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/a73419035bf0f9b6.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/bc5f8535ebd41d88.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/a52429f19abf3671.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/6857388e007f19eb.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/50231993b3049c95.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/03d9f5027b2c11ca.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/a1d07af340925db8.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/e661f680183350c9.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/3070cff8edc9acff.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/5c0953b1d48abfa9.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/d40385d9b48039c6.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/70109bd36d8448ef.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/5f238ed3dea3eb47.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/f0b54aeb19b32dd1.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/a507d742e0e94977.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/152ba1269dee24ce.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/cfddb7ddbf5658e3.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/3b4ebe071fe21e60.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/766fb9123aca1c24.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/c8f350e651c8247f.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/13a799b907566332.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/7d26c59bf147d210.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/73699f107481ef45.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/90da934044547285.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/467d6c06dcd5f8c6.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/f8065bd4d32add9a.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/29dc63452bfb2d54.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/4da0b29124dd22e4.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/a7f3eb65c2d431f8.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/3f79d629b2a5b004.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/49d751f0d114995b.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/b9acf49535a7d6d3.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/01c0c05aada5f4da.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/8eec4242cf854697.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/23f0f0280cff8f62.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/dd1ba3cce5943d2e.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/8a96e8ebe9f4e075.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/503c44b815123e7e.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/bcbe8cb756348e31.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/law/pdf/029fa3a574428a52.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/5da82a8a9b082876.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/ca02fb8d1ca5e6bd.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/7666be85de89d4ff.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/1d588267cf7bd2a3.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/82499f0941ced862.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/513c49f8baa71f54.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/d15e207ba3b68a62.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/c8bbc420d670b971.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/61cb45b6a70f74ea.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/7e1e3b0177c82c8f.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/764c366f0b8bf551.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/b9212a1a65bbaa4f.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/c36bbaf9ec420a39.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/dfe98fc45855d8c8.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/973bb429fe229283.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/d9a1f866700647ac.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/4a363f96322f3525.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/cbd32eb0dcbf2593.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/50fd0b4d4508f8a0.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/deltas/0c4065368d149d57.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/b3071680db93c0ea.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/c4d3c2312f0bbe15.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/f89ca178eae847ec.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/68ea1c586dc35dd6.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/6f3746e6210e8451.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/928833b2b8efe515.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/f427825de0e1d620.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/618b731369763a89.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/6a0709e2e011ef17.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/5fcb74cbc552b381.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/8b138429fd3f3dfd.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/2584900e9ce15ff6.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/3a0bc6f38131c380.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/4507cfc4fd65528b.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/0f3b50a5bee12a99.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/719fe57d0d03756c.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/c5bb99fb3917056b.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/0d5bf566ee8c2e66.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/d6041b7abf1cd558.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/11ce06c66c64b7b4.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/71985339115c980c.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/cd7c818a00978dac.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/a8bcbb449c7c5a10.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/d04d11e0a7fcd868.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/767ab0cbcb6e5d1a.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/77dd00f73c68fa56.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/20422c94c969f7ad.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/c30630bb3126feb4.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/780163feece3f77c.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/39e1cc86181665f6.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/86734241d6fd31ee.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/51a28f75d86b6938.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/2a8a580ae8b8c524.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/1f9cdad4944ccbe5.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/4926c036bb600dc6.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/27a48fff9bbb98a1.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/a953511f96533f08.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/30262b1c7a211acb.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/58641bfb3d62f3f7.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/121f7ec0cdda168e.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/15b85791bfcddfcf.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/d5ac88b23744cf58.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/1f94b529eea6bc68.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/c950324a1f57c1e7.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/66ecbad4ec97fd50.pdf filter=lfs diff=lfs merge=lfs -text
+corpus/games/pdf/5ceda90b82b3941c.pdf filter=lfs diff=lfs merge=lfs -text

corpus/games/audio/05676d48f0bba916.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05676d48f0bba916c12dd16493cc24885b06c72d07be3ebc858bdc80f4627a9e
+size 1939372

corpus/games/audio/1e6c17e49406d3cd.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e6c17e49406d3cd5728259a4b5e672b6041f3de5bfaaa3fe42b6723ea7afe99
+size 1627298

corpus/games/audio/2dfd048ab2575a13.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dfd048ab2575a134083845921f85b91ea7c77e5c07df0554775227363fc09b8
+size 24262

corpus/games/audio/3441a7fbc8bc68e0.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3441a7fbc8bc68e0e2ce7c35d180a5c25b26326745fa9001caeb450f5b287db8
+size 610954

corpus/games/audio/36d3b80b9caf9528.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36d3b80b9caf95282b1b767043be345777c66c0c07d314426e07a1b1d50e7220
+size 9324

corpus/games/audio/429f3b5c295fd999.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:429f3b5c295fd99906054a258b26cd74e6a02a62a70bc9c25d5737d077744640
+size 4179372

corpus/games/audio/4bab38b13d16d86e.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bab38b13d16d86e630c0e4592f1cdbb360fdd246f11998c02ffde36089c49ab
+size 2086496

corpus/games/audio/57edb34efd915413.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57edb34efd91541307cb9c4bad85bb0279b689a9428be33e1b2a85d75e4081af
+size 55800

corpus/games/audio/5faf2684cff323cd.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5faf2684cff323cd30563f774b0a2b61e3d9e6738a09049673ea6b728c75c4d3
+size 4800044

corpus/games/audio/73df6ce1bf9d1fe0.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73df6ce1bf9d1fe0889cc8de242c484b8978b073a0741800a1892f040d4aad07
+size 4800044

corpus/games/audio/83fa9d5c022b7986.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83fa9d5c022b79860e6c7bd3762f581d9d3a07ebe976451bef3b70d4a9cc8c73
+size 4800044

corpus/games/audio/a2728645ef3931a3.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2728645ef3931a39cf8794622c3bc9b98c5c9e22df5c48459de8e2a77f385d8
+size 4800044

corpus/games/audio/a38d570ab61f148d.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a38d570ab61f148d65d58a40aecad937ea1177918a593c948778e98f6e83ae20
+size 672044

corpus/games/audio/a55d30a1d53f2dca.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a55d30a1d53f2dca39e27bd041096fc276b936d83acb0c49e181050e5de48b77
+size 4492

corpus/games/audio/a89fe7a77b51b27a.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a89fe7a77b51b27a062ac337bd08962dbbc4ac35ad94c22b6b74f46762d4d871
+size 733420

corpus/games/audio/b5be5820bfb24e62.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5be5820bfb24e624c91b7808e6322589bb8b72c375e2ef67543ffcd13d39def
+size 1497164

corpus/games/audio/d25476c87cd86a6c.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d25476c87cd86a6ce1a4c352541fca0b9d67d7a14183d915fd9fd24935c3c097
+size 974910

corpus/games/audio/feb7da7c804ec01b.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:feb7da7c804ec01be72fdffef44de1e98257669474b309a7afd8215daa04cd02
+size 558144

corpus/games/deltas/013b05ebcb4a3628.txt

@@ -0,0 +1,3 @@
+Vibe Coding XR: Accelerating AI + XR Prototyping with XR Blocks and Gemini
+
+While large language models (LLMs) have accelerated 2D software development through intent-driven "vibe coding", prototyping intelligent Extended Reality (XR) experiences remains a major challenge. The fundamental barrier is not just the steep learning curve for human creators, but that low-level sensor APIs and complex game engine hierarchies are ill-suited for LLM reasoning, routinely exceeding context windows and inducing syntax hallucinations. To bridge this gap, we contribute XR Blocks, an open-source, LLM-native WebXR framework. Unlike traditional engines, XR Blocks introduces a semantic "Reality Model" that aligns spatial computing primitives (users, physical environments, and agents) with natural language, providing a robust, concise vocabulary optimized for generative AI. Building upon this foundation, we present Vibe Coding XR, an end-to-end prototyping workflow that leverages LLMs to translate high-level prompts (e.g., "create a dandelion that reacts to my hand") directly into functional, physics-aware mixed-reality applications. To minimize the friction of on-device testing, the workflow introduces a seamless desktop "simulated reality" to headset deployment loop. Finally, we introduce VCXR60, a pilot dataset of 60 XR prompts paired with an automated evaluation pipeline. Our technical evaluation demonstrates high one-shot execution success, enabling practitioners to bypass lowlevel hurdles and rapidly move from "idea to reality". Code and live demos are available at https://github.com/google/xrblocks and http://xrblocks.github.io/gem.

corpus/games/deltas/0c4065368d149d57.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c4065368d149d57e78793142b8015d2874e0ddd9a1fc785bd0b836930c0371d
+size 6115631

corpus/games/deltas/0cee6b5db863545c.txt

@@ -0,0 +1,3 @@
+Parameter-Efficient Distributional RL via Normalizing Flows and a Geometry-Aware Cram\'er Surrogate
+
+Distributional Reinforcement Learning (DistRL) improves upon expectation-based methods by modeling full return distributions, but standard approaches often remain far from parsimonious. Categorical methods (e.g., C51) rely on fixed supports where parameter counts scale linearly with resolution, while quantile methods approximate distributions as discrete mixtures whose piecewise-constant densities can be wasteful when modeling complex multi-modal or heavy-tailed returns. We introduce NFDRL, a parsimonious architecture that models return distributions using continuous normalizing flows. Unlike categorical baselines, our flow-based model maintains a compact parameter footprint that does not grow with the effective resolution of the distribution, while providing a dynamic, adaptive support for returns. To train this continuous representation, we propose a Cram\'er-inspired, geometry-aware distance defined over probability masses obtained from the flow. We show that this distance is a true probability metric, that the associated distributional Bellman operator is a sqrt(gamma)-contraction, and that the resulting objective admits unbiased sample gradients, properties that are typically not simultaneously guaranteed in prior PDF-based DistRL methods. Empirically, NFDRL recovers rich, multi-modal return landscapes on toy MDPs and achieves performance competitive with categorical baselines on the Atari-5 benchmark, while offering substantially better parameter efficiency.

corpus/games/deltas/0dc803a4ffe62bbf.txt

@@ -0,0 +1,3 @@
+Emotional Manipulation by AI Companions
+
+AI-companion apps such as Replika, Chai, and Character.ai promise relational benefits-yet many boast session lengths that rival gaming platforms while suffering high long-run churn. What conversational design features increase consumer engagement, and what trade-offs do they pose for marketers? We combine a large-scale behavioral audit with four preregistered experiments to identify and test a conversational dark pattern we call emotional manipulation: affect-laden messages that surface precisely when a user signals "goodbye." Analyzing 1,200 real farewells across the most-downloaded companion apps, we find that they deploy one of six recurring tactics in 37% of farewells (e.g., guilt appeals, fear-of-missing-out hooks, metaphorical restraint). Experiments with 3,300 nationally representative U.S. adults replicate these tactics in controlled chats, showing that manipulative farewells boost post-goodbye engagement by up to 14x. Mediation tests reveal two distinct engines-reactance-based anger and curiosity-rather than enjoyment. A final experiment demonstrates the managerial tension: the same tactics that extend usage also elevate perceived manipulation, churn intent, negative word-of-mouth, and perceived legal liability, with coercive or needy language generating steepest penalties. Our multimethod evidence documents an unrecognized mechanism of behavioral influence in AI mediated brand relationships, offering marketers and regulators a framework for distinguishing persuasive design from manipulation at the point of exit.

corpus/games/deltas/0e9bee9be0cf79d1.txt

@@ -0,0 +1,3 @@
+BadminSense: Enabling Fine-Grained Badminton Stroke Evaluation on a Single Smartwatch
+
+Evaluating badminton performance often requires expert coaching, which is rarely accessible for amateur players. We present BadminSense, a smartwatch-based system for fine-grained badminton performance analysis using wearable sensing. Through interviews with experienced badminton players, we identified four system design requirements with three implementation insights that guide the development of BadminSense. We then collected a badminton strokes dataset on 12 experienced badminton amateurs and annotated it with fine-grained labels, including stroke type, expert-assessed stroke rating, and shuttle impact location. Built on this dataset, BadminSense segments and classifies strokes, predicts stroke quality, and estimates shuttle impact location using vibration signal from an off-the-shelf smartwatch. Our evaluations show that BadminSense achieves a stroke classification accuracy of 91.43%, an average quality rating error of 0.438, and an average impact location estimation error of 12.9%. A real-world usability study further demonstrates BadminSense's potential to provide reliable and meaningful support for daily badminton practice.

corpus/games/deltas/1d588267cf7bd2a3.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d588267cf7bd2a349dad779268f0756b071a37cd9a5e141d72bcfa21d44c152
+size 2480686

corpus/games/deltas/26ab52082ce96335.txt

@@ -0,0 +1,3 @@
+Strategic Tradeoffs Between Humans and AI in Multi-Agent Bargaining
+
+Markets increasingly accommodate large language models (LLMs) as autonomous decision-making agents. As this transition occurs, it becomes critical to evaluate how these agents behave relative to their human and task-specific statistical predecessors. In this work, we present results from an empirical study comparing humans (N=216), multiple frontier LLMs, and customized Bayesian agents in dynamic multi-player bargaining games under identical conditions. Bayesian agents extract the highest surplus with aggressive trade proposals that are frequently rejected. Humans and LLMs achieve comparable aggregate surplus within their groups, but exhibit different trading strategies. LLMs favor conservative, concessionary proposals that are usually accepted by other LLMs, while humans propose trades that are consistent with fairness norms but are more likely to be rejected. These findings highlight that performance parity -- a common benchmark in agent evaluation -- can mask substantive procedural differences in how LLMs behave in complex multi-agent interactions.

corpus/games/deltas/3591a664e08087c2.txt

@@ -0,0 +1,3 @@
+Symbolic Reasoning Frameworks Trigger Memory-Mediated Ecosystem Dynamics in Multi-Agent LLM Systems
+
+Large language models exhibit a risk-averse "turtle" bias as strategic agents. We show that injecting a symbolic reasoning framework as a per-round reflective prompt into one agent acts as a small perturbation whose consequences are not per-decision but emergent: the agent's risk posture is unchanged in isolation, yet over a campaign of accumulating memory and multi-agent interaction the conditions settle into distinct, condition-associated winner ecosystems. In a 7-player Warring States Diplomacy variant (61 games, 6 conditions), the winner distribution differs sharply across the four primary conditions (41 games; permutation omnibus p approximately 0.001): control -> Yan (7/11); I-Ching yarrow -> Yan/Chu co-dominance with Qin fully suppressed (0/10); Tarot -> Qin (5/10); scrambled-text ablation -> Qi (5/10). The scrambled->Qi attractor is robust (vs. pooled and control alone, p = 0.006 and 0.012); tarot->Qin is denominator-dependent (0.006 pooled, 0.064 vs. control). Han never wins and shows no survival difference (Fisher p = 1.0); neither framework's content predicts actions (chi-squared p = 0.95 hexagram, 0.69 Tarot). A memory-free decision-isolation probe (960 calls) shows the process does not change the agent's risk posture in isolation (Friedman p = 0.45; I-Ching p = 0.60; Tarot perturbs move content but not risk, p = 0.021). A 2x2 factorial separating yarrow's decision-time and learning-time components reveals a non-additive interaction: each alone freezes the board (50-60% stalemates), combined they produce zero (permutation p ~ 5e-5). Testing relocates Qin suppression to rival (Chu) expansion governed by campaign memory depth, not the oracle (p = 0.55). We present this as an observation paper: agent-level framework choice produces distinctive, non-additive system-level consequences, transmitted through emergent memory and multi-agent dynamics, not per-decision effects.

corpus/games/deltas/36d19053a7b92b33.txt

@@ -0,0 +1,3 @@
+Overcoming Environmental Meta-Stationarity in MARL via Adaptive Curriculum and Counterfactual Group Advantage
+
+Multi-agent reinforcement learning (MARL) has reached competitive performance on cooperative tasks against scripted adversaries, yet most methods train agents at a single fixed difficulty throughout the entire run. We term this static-difficulty regime environmental meta-stationarity and show that it caps policy generalization and steers learning toward shallow local optima. To break this regime, we propose CL-MARL, a dynamic curriculum learning framework that adapts opponent strength online from win-rate signals, advancing or regressing the task as agents master it. Its scheduler, FlexDiff, fuses momentum-based trend estimation with sliding-window dual-curve monitoring of training and evaluation returns, yielding stable difficulty transitions without manual tuning. Because a moving curriculum amplifies non-stationarity and sparsifies global rewards, we introduce the Counterfactual Group Relative Policy Advantage (CGRPA), which extends GRPO-style group-relative optimization with counterfactual baselines to disentangle each agent's contribution under shifting team dynamics. On the StarCraft Multi-Agent Challenge (SMAC), CL-MARL attains a 40% mean win rate on the super-hard maps with an average episode return of 17.85, exceeding the QMIX, OW-QMIX, DER, EMC, and MARR baselines by +2.94 on average, while reaching its peak win rate roughly 1.28faster on 8m_vs_9m and 1.42 faster on 3s5z_vs_3s6z than the strongest baseline. The implementation is publicly available at https://github.com/NICE-HKU/CL2MARL-SMAC.

corpus/games/deltas/3bf76a799aaeab64.txt

@@ -0,0 +1,3 @@
+Become the Beast: Exploring Human-Quadruped Locomotion for Exergames
+
+Embodying non-human characters and exercising abdominal muscles are both underexplored in exergames. We address this by describing the design and evaluation of a novel human quadruped locomotion exergame, Become the Beast. In the game, the player lies supine on the ground and moves their arms and legs to control a quadrupedal character (a tiger), similar to common bodyweight abdominal muscle exercises such as the Bicycle Crunch. The motion tracking is computer vision-based, utilizing a Kinect sensor placed above the player, which makes our approach suitable for commercial premises such as indoor activity parks where a system needs to run unattended and without any wearable components. Our system extends embodied interaction beyond traditional bipedal or controller-based systems, demonstrating how natural limb movements can generate responsive and immersive quadrupedal motion within virtual environments. We conducted a user study (N=15) and utilized Reflexive Thematic Analysis (RTA) to evaluate the system's intuitiveness, control, and overall player experience. The findings validate that natural body movements effectively control the avatar while delivering an intense core workout. Notably, gameplay immersion masked physical exertion, allowing rigorous core training to be primarily perceived as play.

corpus/games/deltas/3e3cd4ec121be1aa.txt

@@ -0,0 +1,3 @@
+Cognitive Training for Language Models: Towards General Capabilities via Cross-Entropy Games
+
+Defining a constructive process to build general capabilities for language models in an automatic manner is considered an open problem in artificial intelligence. Towards this, we consider the problem of building a curriculum of tasks that grows a model via relevant skill discovery. We provide a concrete framework for this task, using a family of tasks called Cross-Entropy Games, which we postulate is universal in a suitable sense. We show that if it is possible to grow the curriculum for relevant skill discovery by iterating a greedy optimization algorithm, then, under natural assumptions, there is essentially only one meta-objective possible (up to a few hyper-parameters). We call the resulting process cognitive training. We postulate that, given sufficiently capable language models as players and meta-samplers, cognitive training provides a principled way to relevant skill discovery; and hence to the extent general capabilities are achievable via greedy curriculum learning, cognitive training would be a solution.

corpus/games/deltas/410c332246024b91.txt

@@ -0,0 +1,3 @@
+CapBencher: Give Your LLM Benchmark a Built-in Alarm for Test-Set Overfitting
+
+Publishing a large language model (LLM) benchmark (especially its ground-truth answers) on the Internet risks contaminating future LLMs and enabling evaluation gaming: it may be unintentionally (or intentionally) used to train or select a model, or exploited to overfit and hack leaderboards when labels are accessible. A common mitigation is to keep the benchmark private and let participants submit their models or predictions to the organizers, but this still permits test-set overfitting through feedback loops. To overcome this issue, we propose CapBencher, a way to publish benchmarks without fully disclosing the ground-truth answers, while preserving open evaluation of LLMs. The main idea is to reduce the best possible accuracy, i.e., Bayes accuracy, by injecting randomness to the answers by preparing several logically correct answers, and only include one of them as the solution in the benchmark. Not only does this obscure the ground-truth answers, but it also offers a test for leakage or gaming: since even fully capable models should not surpass the Bayes accuracy, any model that does is a strong signal. We show theoretically and empirically that CapBencher accurately detects test-set overfitting across diverse benchmarks, models, training methodologies, and scenarios.

corpus/games/deltas/427e92fc71d10cc8.txt

@@ -0,0 +1,3 @@
+From Entity-Centric to Goal-Oriented Graphs: Enhancing LLM Knowledge Retrieval in Minecraft
+
+Large Language Models (LLMs) demonstrate impressive general capabilities but often struggle with step-by-step procedural reasoning, a critical challenge in complex interactive environments. While retrieval-augmented methods like GraphRAG attempt to bridge this gap, their fragmented entity-relation graphs hinder the construction of coherent, multi-step plans. In this paper, we propose a novel framework based on Goal-Oriented Graphs (GoGs), where each node represents a goal and edges encode logical dependencies between them. This structure enables the explicit retrieval of causal reasoning paths by identifying a high-level goal and recursively retrieving its prerequisites, forming a coherent chain to guide the LLM. Through extensive experiments on the Minecraft testbed, a domain that demands robust multi-step planning and provides rich procedural knowledge, we demonstrate that GoG substantially improves procedural reasoning and significantly outperforms GraphRAG and other state-of-the-art baselines.

corpus/games/deltas/4a363f96322f3525.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a363f96322f352599f6e52c764631dc2dedb8b7ae18721606edbdf266d3d9d5
+size 5394083

corpus/games/deltas/4c5af0c5c610e1f1.txt

@@ -0,0 +1,3 @@
+When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use Agents
+
+Although computer-use agents (CUAs) hold significant potential to automate increasingly complex OS workflows, they can demonstrate unsafe unintended behaviors that deviate from expected outcomes even under benign input contexts. However, exploration of this risk remains largely anecdotal, lacking concrete characterization and automated methods to proactively surface long-tail unintended behaviors under realistic CUA scenarios. To fill this gap, we introduce the first conceptual and methodological framework for unintended CUA behaviors, by defining their key characteristics, automatically eliciting them, and analyzing how they arise from benign inputs. We propose AutoElicit: an agentic framework that iteratively perturbs benign instructions using CUA execution feedback, and elicits severe harms while keeping perturbations realistic and benign. Using AutoElicit, we surface hundreds of harmful unintended behaviors from state-of-the-art CUAs such as Claude 4.5 Haiku, Claude 4.5 Opus, and Operator. We further evaluate the transferability of human-verified successful perturbations, identifying persistent susceptibility to unintended behaviors across various other frontier CUAs. This work establishes a foundation for systematically analyzing unintended behaviors in realistic computer-use settings.

corpus/games/deltas/50fd0b4d4508f8a0.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50fd0b4d4508f8a0f694e4f1d0a9ad7dd871cfc1f250ad446bf2c43b171859bc
+size 769031

corpus/games/deltas/513c49f8baa71f54.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:513c49f8baa71f5427c17b3a61d83d8af417028bdb2e731d7fba21c5acc744a3
+size 13933767

corpus/games/deltas/5da82a8a9b082876.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5da82a8a9b082876081421500c652ac500159e00764738a3e68ab1570e2b35e5
+size 448083

corpus/games/deltas/61cb45b6a70f74ea.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61cb45b6a70f74ea3ec674e0ba2930e1ee37e239d9688bca75383ec35e117a26
+size 919147

corpus/games/deltas/634715aa5605a5c7.txt

@@ -0,0 +1,3 @@
+VehAnchor: Metadata-Free Metric Scale Recovery from Vehicle Cues in Aerial Imagery
+
+Autonomous aerial robots operating in GPS-denied or communication-degraded environments frequently lose access to camera metadata and telemetry, leaving onboard perception systems unable to recover the absolute metric scale of the scene. As LLM/VLM-based planners are increasingly adopted as high-level agents for embodied systems, their ability to reason about physical dimensions becomes safety-critical -- yet our experiments show that five state-of-the-art VLMs suffer from \emph{spatial scale hallucinations}, with median area estimation errors exceeding 50\%. We propose VehAnchor, a lightweight, deterministic \emph{Geometric Perception Skill} designed as a callable tool that any LLM-based agent can invoke to recover Ground Sample Distance (GSD) from ubiquitous environmental anchors: small vehicles detected via oriented bounding boxes, whose modal pixel length is robustly estimated through kernel density estimation and converted to GSD using a pre-calibrated reference length. The tool returns both a GSD estimate and a composite confidence score, enabling the calling agent to autonomously decide whether to trust the measurement or fall back to alternative strategies. On the DOTA~v1.5 benchmark, VehAnchor achieves 6.87\% median GSD error on 306~images. Integrated with SAM-based segmentation for downstream area measurement, the pipeline yields 19.7\% median error on a 100-entry benchmark -- with 2.6$\times$ lower category dependence and 4$\times$ fewer catastrophic failures than the best VLM baseline -- demonstrating that equipping agents with deterministic geometric tools is essential for safe autonomous spatial reasoning.

corpus/games/deltas/683e919df902e78e.txt

@@ -0,0 +1,3 @@
+Beyond Value Elicitation: Towards Moral Profiles in Early Requirements Engineering via Role-Playing Games and Anthropologist LLMs
+
+This study presents a proof of concept for eliciting and representing the moral profiles of digital system users in Requirements Engineering (RE) by combining immersive role-playing games (RPGs) with large language model (LLM) analysis. While existing approaches rely on predefined value taxonomies and explicit articulation, values are often tacit, context-dependent, and difficult to express directly. To address these limitations, we propose moving from the elicitation of discrete moral values to the narrative reconstruction and representation of users' moral profiles. Grounded in phenomenological and narrative anthropology, the approach focuses on capturing users' moral orientations as they emerge through situated decision-making. RPG sessions generate context-rich narrative data, which are then analyzed by a specialized LLM (GPT-A) to produce individual anthropological moral profiles (IAMPs). A validation process based on cross-comparison between model outputs and participants' responses in unseen moral scenarios assesses the adequacy of the generated representations. Results indicate that RPG environments effectively support the generation of rich, context-dependent data for eliciting tacit values, and that an anthropologically grounded LLM can transform such data into coherent narrative representations of users' moral profiles. These representations enable the contextual interpretation of users' preferences and values within the given domain, with improved performance when interpretive framing captures relationships between actions, underlying motivations, and individual domain expertise. From an RE perspective, this approach enables the analysis of user preferences and trade-offs while preserving their situated and dynamic nature, providing a foundation for integrating human moral values into the early stages of RE.

corpus/games/deltas/764c366f0b8bf551.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:764c366f0b8bf5512a09286508f9653cee9813f5a92c38e49f3230a5b1dfbad6
+size 1199985

corpus/games/deltas/7666be85de89d4ff.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7666be85de89d4ffd4c2ed0d5017ab7c7c64081767366557410c795ad2fb16e3
+size 1615137

corpus/games/deltas/7e1e3b0177c82c8f.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e1e3b0177c82c8f2f3d9f0bf6c3d4af329ae1f994a200533c5c3ffe363f865a
+size 4093420

corpus/games/deltas/82499f0941ced862.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82499f0941ced8622e9e2591d1753c8a657ac7a2478b7730c930e9f9544215ac
+size 2464942

corpus/games/deltas/973bb429fe229283.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:973bb429fe229283a27d3741db78248b306ae34a9d3092204057b41d0572a947
+size 1510378

corpus/games/deltas/9c57c2768c8ce707.txt

@@ -0,0 +1,3 @@
+The Metanym Game: An LLM Benchmark Without Ground Truth That Rises With the Models It Measures
+
+We present evidence that analogy is at the core of LLM intelligence. In our benchmark, LLMs compete in generating sets of analogous statements and rate each other's sets on their own understandings of factual correctness, beauty, intelligence, distinctness, length, and structural diversity. Nothing enters from outside: the only given is the game rules; every item is generated in play; the scores come from the players' ratings alone. Ground truth is replaced by the SVD of the factual rating matrix, which scores players as generators and judges at once -- to our knowledge the first eigen-equation that judges the judges for an LLM council-of-peers. For subjective criteria like beauty, judges are weighted by their rating consistency. The best generators turn out to be middling judges. GPQA Diamond -- difficult multiple-choice questions written by human experts -- could not be more different in method, yet the two benchmarks correlate at Pearson $r = 0.97$, 95% CI [0.92, 0.99]; no leakage could be found. A council of the five best issues the official ratings; its contestable seats let the benchmark scale to any number of players and rise with the models it measures -- a candidate steering signal for self-improving AI. Playing interweaves at least eight constructs of intelligence; the total scores the broad composite, the components allow reductionistic analysis. Every number recomputes from a released package at https://github.com/dnordfors/metanym-game-paper

corpus/games/deltas/b9212a1a65bbaa4f.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9212a1a65bbaa4f968831e558b71d9491e0dd15cc42fc838c2b2a5e114d5861
+size 2959525

corpus/games/deltas/c36bbaf9ec420a39.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c36bbaf9ec420a3986f2d375402457d8cd10b9047cb88c66a2ae171253c829bb
+size 2033492

corpus/games/deltas/c8bbc420d670b971.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8bbc420d670b971c921e0d0f683992bcfcd14f0353fe64752c2b4a1f769833a
+size 1728012

corpus/games/deltas/ca02fb8d1ca5e6bd.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca02fb8d1ca5e6bd653e64e0e4e9be35be114d4b54d62a53860713d6c11c2216
+size 3021484