{"total":79,"items":[{"citing_arxiv_id":"2606.29178","ref_index":7,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Selective Memory Retention for Long-Horizon LLM Agents","primary_cat":"cs.AI","submitted_at":"2026-06-28T03:46:46+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"TraceRetain applies feature-based scoring to evict low-value entries from bounded external memory in frozen LLM agents, preserving task success under 75% synthetic distractors on ALFWorld where unbounded memory degrades.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.05420","ref_index":3,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Assessing the Carbon Emissions and Energy Consumption of U.S. Hyperscale Data Centers","primary_cat":"cs.AI","submitted_at":"2026-06-03T20:38:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"US hyperscale data centers consumed 68-99 TWh electricity and emitted 37-54 Mt CO2, representing 1.8% of US electricity use with average carbon intensity 48% above the national grid average.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.04552","ref_index":7,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"LDARNet: DNA Adaptive Representation Network with Learnable Tokenization for Genomic Modeling","primary_cat":"cs.CL","submitted_at":"2026-06-03T07:38:17+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"LDARNet learns adaptive token boundaries via dynamic chunking in a genomic foundation model and reports gains on histone modification tasks over larger models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.01766","ref_index":5,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Night-Window Batching versus Carbon-Aware Scheduling for Clinical AI GPU Workloads","primary_cat":"cs.DC","submitted_at":"2026-06-01T06:49:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Simulation finds overnight batching captures 78% of carbon reduction from mixed carbon-urgency scheduling while missing fewer urgent deadlines on an 8-GPU setup.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.02643","ref_index":29,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Inference Cost Attacks for Retrieval-Augmented Large Language Models","primary_cat":"cs.CR","submitted_at":"2026-05-31T15:11:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Poisoning external knowledge bases with LLM-agent-crafted documents can increase RAG inference token consumption by up to 13.12 times at over 90% success rate while preserving answer quality.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.07632","ref_index":161,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Evaluation of ML Resource Utilization Requires Model Life Cycle Assessment","primary_cat":"cs.LG","submitted_at":"2026-05-31T05:58:12+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"The paper calls for life cycle assessment to capture embodied hardware costs and full pipeline operational costs in AI development and deployment.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00811","ref_index":11,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Certificates without Electrons? Theory and Evidence on Impacts from AI-Driven Power Demand","primary_cat":"econ.EM","submitted_at":"2026-05-30T16:57:05+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Game-theoretic modeling and difference-in-differences analysis using LLM releases show AI data center demand increases fossil generation, wholesale prices, and outages near data centers unless mitigated by behind-the-meter capacity.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.05216","ref_index":211,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"A Comprehensive Survey on Semantic Communication in Non-Terrestrial Networks: Architectures, Methodologies, and Challenges","primary_cat":"cs.IT","submitted_at":"2026-05-28T09:02:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"A literature survey that pairs NTN limitations with semantic communication properties, organizes work by platform and methodology, and lists open problems for integrated SAGIN systems.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.26848","ref_index":3,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Design principles for optoelectronic light-scattering reservoir computing at the edge of chaos","primary_cat":"physics.optics","submitted_at":"2026-05-26T11:01:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Authors map reservoir dynamics, input coupling, and interconnectivity in a light-scattering optoelectronic reservoir to locate quantitative optima that yield stable Mackey-Glass prediction and 84.5% accuracy on spoken-digit classification.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.25966","ref_index":20,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Mapping the Schedule x Bit-Width Boundary in Sub-100M Quantisation-Aware Training","primary_cat":"cs.LG","submitted_at":"2026-05-25T15:42:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Factorial experiments with over 1300 runs falsify the hypothesis that INT6 QAT needs a different LR schedule from higher precision and identify a 50M-parameter boundary for INT4 schedule sensitivity.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.24124","ref_index":10,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"A generative pre-trained transformer with Kerr-soliton attention","primary_cat":"physics.optics","submitted_at":"2026-05-22T18:34:06+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Kerr-soliton attention realizes transformer attention in physical hardware via Kerr solitons in a resonator, with analytic training and experimental inference showing high-fidelity agreement between hardware and model.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.23051","ref_index":7,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"General-Purpose Photonic Computing Primitive for Contemporary Artificial Intelligence","primary_cat":"physics.optics","submitted_at":"2026-05-21T21:33:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DUET is a photonic tensor core paradigm that uses structural symmetry in VODICs to support arbitrary signed operands directly, experimentally tested on image classification, segmentation, and Transformer tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.21847","ref_index":25,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"CompPow: A Case for Component-level GPU Power Management","primary_cat":"cs.AR","submitted_at":"2026-05-21T00:44:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"CompPow makes the case that component-aware power management inside GPUs can yield 10% higher energy efficiency and 5% better performance for ML workloads.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.22883","ref_index":32,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Energy per Successful Goal: Goal-Level Energy Accounting for Agentic AI Systems","primary_cat":"cs.AI","submitted_at":"2026-05-20T22:55:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Proposes EpG and OOI metrics showing agentic workflows use 4.33x more energy per successful goal than linear baselines due to orchestration structure.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.21322","ref_index":56,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Optimized Federated Knowledge Distillation with Distributed Neural Architecture Search","primary_cat":"cs.LG","submitted_at":"2026-05-20T15:50:49+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"FedKDNAS combines client-side neural architecture search with knowledge distillation from aggregated server predictions to improve accuracy and efficiency in heterogeneous federated learning.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20815","ref_index":12,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"GraphRAG on Consumer Hardware: Benchmarking Local LLMs for Healthcare EHR Schema Retrieval","primary_cat":"cs.CL","submitted_at":"2026-05-20T07:09:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"GraphRAG with 7-8B local LLMs on 8GB VRAM hardware builds knowledge graphs from EHR docs and answers queries, with Llama 3.1 creating the largest graph, Qwen 2.5 scoring highest on quality, and models below ~7B failing to complete the pipeline.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20281","ref_index":20,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"The Economics of AI Inference: Inflation Dynamics, Welfare Costs, and Optimal Monetary Policy under the Inference-Cost Phillips Curve","primary_cat":"econ.GN","submitted_at":"2026-05-19T05:26:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Develops the Inference-Cost Phillips Curve linking AI inference costs to inflation dynamics, derives structural slopes and optimal monetary policy, and reports empirical estimates from US and G7 data that align with theoretical predictions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.19195","ref_index":67,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"The Thermodynamic Costs of Simple Linear Regression","primary_cat":"cond-mat.stat-mech","submitted_at":"2026-05-18T23:51:02+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Thermodynamic lower bounds are approximated for exact and SGD linear regression, producing energy-aware scaling laws for optimal training dataset size given a target generalization error.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16844","ref_index":14,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Artificial Adaptive Intelligence: The Missing Stage Between Narrow and General Intelligence","primary_cat":"cs.AI","submitted_at":"2026-05-16T07:04:28+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Proposes Artificial Adaptive Intelligence as the regime between narrow and general AI, defined by elimination of human-specified hyperparameters, and introduces an adaptivity index plus parametric minimality principle grounded in minimum description length.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16538","ref_index":4,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"LLMs in Qualitative Research: Opportunities, Limitations, and Practical Considerations","primary_cat":"cs.HC","submitted_at":"2026-05-15T18:33:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"The paper outlines opportunities, limitations, and practical parameters for integrating LLMs into qualitative research while aligning with epistemological commitments like reflexivity and interpretive judgment.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.15551","ref_index":53,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Characterizing Learning in Deep Neural Networks using Tractable Algorithmic Complexity Analysis","primary_cat":"cs.LG","submitted_at":"2026-05-15T02:44:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"QuBD extends algorithmic complexity estimation to quantized DNN weights, revealing that complexity decreases during learning, increases with overfitting, follows grokking patterns, and correlates with generalization.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Therefore, the residual loss at precisionqis the information needed to refiney q into the target object x⋆, up to the same logarithmic term. As more bit-planes are retained, this missing information decreases; at full-precision only the additive constant remains. Repeated application of Thm. 3.2 gives the equivalent telescoping sum Rq −R r = r−1X j=q C(bj+1 |y j) +O((r−q) log(q ⋆d)),(53) which shows that one conditional contribution is removed at each added bit-plane. C.4 Empirical Validation of the Decrease in Residual Loss We test the estimation gap using the CTM table as a finite reference. By the coding theorem, CTM assigns low complexity to blocks with high empirical algorithmic probability, up to an additive constant (Solomonoff, 1964; Levin, 1974; Delahaye and Zenil, 2012)."},{"citing_arxiv_id":"2605.15413","ref_index":58,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Transformer Scalability Crisis: The First Comprehensive Empirical Analysis of Performance Walls in Modern Language Models","primary_cat":"cs.LG","submitted_at":"2026-05-14T20:57:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Empirical tests on 118 transformers show success falling from 88.1% at 512 tokens to 0% at 2048 tokens, with compressed models achieving 649.2 tokens/sec/M parameters versus 12.5 for large generative ones.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.14690","ref_index":24,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Integrated photonic computing: towards high-dimensional information processing","primary_cat":"physics.optics","submitted_at":"2026-05-14T11:03:29+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"A review of integrated photonic computing that organizes low- to high-dimensional architectures and argues that exploiting light's full dimensionality offers a path to scalable, energy-efficient information processing.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.14624","ref_index":14,"ref_count":2,"confidence":0.98,"is_internal_anchor":true,"paper_title":"An Amortized Efficiency Threshold for Comparing Neural and Heuristic Solvers in Combinatorial Optimization","primary_cat":"cs.LG","submitted_at":"2026-05-14T09:39:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Introduces the Amortized Efficiency Threshold (AET) to identify the deployment volume at which neural combinatorial optimization solvers achieve lower total energy use than heuristic baselines after accounting for training costs.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.13114","ref_index":10,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Recasting AI Data Centers as Engines for Carbon Removal","primary_cat":"math.OC","submitted_at":"2026-05-13T07:33:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"AI data center waste heat upgraded by heat pumps can drive direct air capture to achieve net CO2 removal and offset operational emissions in several US states under current and 2030 scenarios.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11733","ref_index":10,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Position: LLM Inference Should Be Evaluated as Energy-to-Token Production","primary_cat":"cs.CE","submitted_at":"2026-05-12T08:15:04+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"LLM inference should be reframed and evaluated as energy-to-token production with a Token Production Function that accounts for power, cooling, and efficiency ceilings.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Smith, and O. Etzioni. Green ai.Communications of the ACM, 63 (12):54-63, 2020. doi: 10.1145/3381831. URLhttps://doi.org/10.1145/3381831. [9] D. Patterson, J. Gonzalez, Q. Le, C. Liang, L. M. Munguia, D. Rothchild, J. Dean, et al. Carbon emissions and large neural network training, 2021. URL https://arxiv.org/abs/2104.1 0350. arXiv preprint arXiv:2104.10350. [10] D. Patterson, J. Gonzalez, U. Hölzle, Q. Le, C. Liang, L.-M. Munguia, J. Dean, et al. The carbon footprint of machine learning training will plateau, then shrink.Computer, 55(7):18-28, 2022. doi: 10.1109/MC.2022.3148714. URLhttps://doi.org/10.1109/MC.2022.3148714. [11] C. J. Wu, R. Raghavendra, U. Gupta, B. Acun, N. Ardalani, K. Maeng, K. Hazelwood, et al."},{"citing_arxiv_id":"2605.09060","ref_index":27,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Language-Conditioned Visual Grounding with CLIP Multilingual","primary_cat":"cs.CL","submitted_at":"2026-05-09T17:06:16+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Fixing the visual encoder in multilingual CLIP isolates text-branch deficits as the cause of lower visual grounding performance for low-resource languages, with model scaling widening some gaps but not others.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Awareness of the environmental cost of deep learning has grown rapidly since Strubell et al. [25] quantified the carbon footprint of large transformer training. Schwartz et al. [26] responded with a call forGreen AI, treating computational efficiency as a first-class publication criterion. Subsequent work refined these estimates by accounting for hardware generation and grid carbon intensity [27]-[29], and the focus shifted from training to inference: Luccioni et al. [17] provided the first systematic comparison of inference energy across NLP tasks and architectures, finding that generative models are substantially more expensive than discriminative ones, and that energy per query varies by up to two orders of magni- tude across tasks. The AI Energy Score [16] addresses the"},{"citing_arxiv_id":"2605.08441","ref_index":25,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"DUET: Optimize Token-Budget Allocation for Reinforcement Learning with Verifiable Rewards","primary_cat":"cs.LG","submitted_at":"2026-05-08T20:03:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"DUET improves RLVR by allocating tokens across both prompt selection and rollout length, outperforming full-budget baselines even when using only half the tokens.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"[23] Hieu Trung Nguyen, Bao Nguyen, Wenao Ma, Yuzhi Zhao, Ruifeng She, and Viet Anh Nguyen. Adaptive rollout allocation for online reinforcement learning with verifiable rewards. InInternational Conference on Learning Representations, 2026. arXiv:2602.01601. [24] Art B. Owen.Monte Carlo Theory, Methods and Examples. 2013. Online manuscript, Stanford University. [25] David Patterson, Joseph Gonzalez, Quoc Le, Chen Liang, Lluis-Miquel Munguia, Daniel Rothchild, David So, Maud Texier, and Jeff Dean. Carbon emissions and large neural network training.arXiv preprint arXiv:2104.10350, 2021. [26] David Rein, Betty Li Hou, Asa Cooper Stickland, Jackson Petty, Richard Yuanzhe Pang, Julien Dirani, Julian Michael, and Samuel R."},{"citing_arxiv_id":"2605.07223","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"A Hardware-aware Hopfield Network with a Nonlinear Memristor Array for Robust Associative Memory with Superlinear Capacity","primary_cat":"cond-mat.dis-nn","submitted_at":"2026-05-08T04:16:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A memristor-array Hopfield network uses device nonlinearity to exceed classical memory capacity with K ~ 0.14N experimentally and superlinear K ~ 0.3 N^1.2 in simulations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06597","ref_index":49,"ref_count":2,"confidence":0.98,"is_internal_anchor":true,"paper_title":"UniSD: Towards a Unified Self-Distillation Framework for Large Language Models","primary_cat":"cs.CL","submitted_at":"2026-05-07T17:22:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"UniSD unifies self-distillation components for autoregressive LLMs and its full integrated version improves base models by 5.4 points and baselines by 2.8 points across six benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05519","ref_index":39,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"OpenG2G: A Simulation Platform for AI Datacenter-Grid Runtime Coordination","primary_cat":"cs.LG","submitted_at":"2026-05-06T23:36:16+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"OpenG2G is a new extensible simulation platform that lets users implement and compare classic, optimization, and learning-based controllers for AI datacenter power flexibility coordinated with the grid.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05416","ref_index":75,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"From Cradle to Cloud: A Life Cycle Review of AI's Environmental Footprint","primary_cat":"cs.CY","submitted_at":"2026-05-06T20:20:17+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"A review of AI sustainability studies finds inconsistent life cycle definitions and predominant reliance on coarse CO2e proxies, with limited coverage of water, materials, and multi-impact assessments.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.03667","ref_index":8,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"ELAS: Efficient Pre-Training of Low-Rank Large Language Models via 2:4 Activation Sparsity","primary_cat":"cs.LG","submitted_at":"2026-05-05T12:04:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"ELAS pre-trains low-rank LLMs by applying 2:4 activation sparsity after squared ReLU to cut memory and accelerate training with minimal performance loss.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.02300","ref_index":275,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"A Meta Reinforcement Learning Approach to Goals-Based Wealth Management","primary_cat":"cs.LG","submitted_at":"2026-05-04T07:48:02+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"MetaRL pre-trained on GBWM problems delivers near-optimal dynamic strategies in 0.01s achieving 97.8% of DP optimal utility and handles larger problems where DP fails.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.01793","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Analytic Framework for Estimating Memory Cost","primary_cat":"cs.ET","submitted_at":"2026-05-03T09:17:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"An analytic framework is introduced to estimate memory-related energy costs of AI models and quantify their ecological footprint.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.00315","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Unbox Responsible GeoAI: Navigating Climate Extreme and Disaster Mapping","primary_cat":"cs.CY","submitted_at":"2026-05-01T00:49:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Responsible GeoAI for disaster mapping requires governance across data, applications, and society rather than algorithm improvements alone.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.00300","ref_index":15,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Token Arena: A Continuous Benchmark Unifying Energy and Cognition in AI Inference","primary_cat":"cs.AI","submitted_at":"2026-05-01T00:05:54+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"TokenArena is a continuous benchmark for AI inference endpoints that measures output speed, time to first token, blended price, effective context, quality, and modeled energy to produce composites of joules per correct answer, dollars per correct answer, and endpoint fidelity.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.25903","ref_index":39,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Carbon-Taxed Transformers: A Green Compression Pipeline for Overgrown Language Models","primary_cat":"cs.SE","submitted_at":"2026-04-28T17:48:16+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"CTT is a compression pipeline for LLMs that achieves up to 49x memory reduction, 10x faster inference, 81% lower CO2 emissions, and retains 68-98% accuracy on code clone detection, summarization, and generation tasks.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"models, often containing tens or hundreds of billions of parameters, are capable of producing functionally correct code and even solving competitive programming problems [ 29]. Early studies report promising productivity gains when such AI pair programmers are introduced into real developer workflows, for example, a controlled experiment found that developers using GitHub Copilot completed a coding task 55% faster than those without it [ 39]. This potential of LLMs in SE has fueled a new paradigm of AI-assisted programming, wherein neural models become integral collaborators in software development. Although LLMs demonstrate remarkable performance, they are very large with billions (B) of parameters, incur high operational costs, consume substantial computational memory, require specialized hardware, and suffer from low inference speed."},{"citing_arxiv_id":"2604.24294","ref_index":29,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"AI-Native Autonomous Infrastructure (ANAI): A Formal Framework for the Next General-Purpose Technology","primary_cat":"eess.SY","submitted_at":"2026-04-27T10:27:55+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Introduces ANAI framework with Autonomy Index (AIx), Infrastructure Coupling Coefficient (ICC), and Technological Transition Potential (TTP) to model AI-driven infrastructural transition via nonlinear coevolution and recursive feedback loops.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.24805","ref_index":8,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"minAction.net: Energy-First Neural Architecture Design -- From Biological Principles to Systematic Validation","primary_cat":"cs.LG","submitted_at":"2026-04-27T06:26:36+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Large-scale experiments show architecture performance depends on task type, not universality, and a single-parameter energy penalty reduces computational energy by ~1000x with negligible accuracy cost.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.19342","ref_index":47,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Are Large Language Models Economically Viable for Industry Deployment?","primary_cat":"cs.CL","submitted_at":"2026-04-21T11:25:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Small LLMs under 2B parameters achieve better economic break-even, energy efficiency, and hardware density than larger models on legacy GPUs for industrial tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.16228","ref_index":23,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"TRON: Trainable, architecture-reconfigurable random optical neural networks","primary_cat":"physics.optics","submitted_at":"2026-04-17T16:44:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TRON demonstrates a trainable and reconfigurable optical neural network that combines multi-scattering media with DMD-based matrix multiplication and performs in-situ optimization plus neural architecture search on the optical hardware itself.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05216","ref_index":16,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"SAT: Sequential Agent Tuning for Coordinator Free Plug and Play Multi-LLM Training with Monotonic Improvement Guarantees","primary_cat":"cs.LG","submitted_at":"2026-04-17T01:45:30+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"SAT trains multi-LLM teams with sequential block updates to deliver monotonic gains and plug-and-play model swaps that provably improve performance bounds.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"fully open-source 14b coder at o3-mini level.Notion Blog(2025). [14] Rémi Munos, Tom Stepleton, Anna Harutyunyan, and Marc Bellemare. 2016. Safe and efficient off-policy reinforcement learning.Advances in neural information processing systems29 (2016). [15] OpenAI. 2023. GPT-4 Technical Report.arXiv preprint arXiv:2303.08774(2023). https://arxiv.org/abs/2303.08774 [16] David Patterson, Joseph Gonzalez, Quoc Le, Chen Liang, Lluis-Miquel Munguia, Daniel Rothchild, David So, Maud Texier, and Jeff Dean. 2021. Carbon Emissions and Large Neural Network Training.arXiv preprint arXiv:2104.10350(2021). [17] Qwen, :, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang,"},{"citing_arxiv_id":"2604.10861","ref_index":7,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Training single-electron and single-photon stochastic physical neural networks","primary_cat":"quant-ph","submitted_at":"2026-04-12T23:57:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Single-electron and single-photon stochastic physical neural networks achieve over 97% MNIST test accuracy when trained with empirical outputs in the backward pass using few trials per layer.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.10272","ref_index":19,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"The Phase Is the Gradient: Equilibrium Propagation for Frequency Learning in Kuramoto Networks","primary_cat":"cs.LG","submitted_at":"2026-04-11T16:28:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"In Kuramoto networks at equilibrium, weak nudging makes phase displacement the exact gradient of loss w.r.t. natural frequencies, enabling frequency learning that beats weight learning and resolves convergence via spectral initialization.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.09048","ref_index":9,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Watt Counts: Energy-Aware Benchmark for Sustainable LLM Inference on Heterogeneous GPU Architectures","primary_cat":"cs.DC","submitted_at":"2026-04-10T07:15:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Watt Counts supplies over 5,000 energy measurements across 50 LLMs and 10 GPUs and shows that hardware-aware selection can reduce server-scenario energy use by up to 70 percent with little effect on user experience.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.06375","ref_index":9,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"SymptomWise: A Deterministic Reasoning Layer for Reliable and Efficient AI Systems","primary_cat":"cs.AI","submitted_at":"2026-04-07T19:00:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"SymptomWise uses expert knowledge and deterministic rules for diagnosis after LLM-based symptom extraction, achieving 88% top-5 accuracy on 42 challenging pediatric neurology cases.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.04096","ref_index":19,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Toward a Sustainable Software Architecture Community: Evaluating ICSA's Environmental Impact","primary_cat":"cs.SE","submitted_at":"2026-04-05T12:29:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"The study provides exploratory estimates of carbon emissions from GenAI inference in ICSA papers and from the full operations of the ICSA 2025 conference.","context_count":1,"top_context_role":"background","top_context_polarity":"support","context_text":"GenAI offers clear benefits, its energy implications warrant careful consideration. B. Carbon Cost of GenAI GenAI's carbon footprint arises primarily fromtrainingand inference. Training large models is energy-intensive and can generate substantial CO 2eq emissions [2]. For example, train- ing GPT-3 required approximately 1,287 MWh of electricity and emitted an estimated 552 tons of CO 2eq [19]. In software architecture research, however, models are typically accessed via APIs or pre-trained checkpoints, making inference the relevant stage for footprint estimation. Although inference consumes less energy per request, its cumulative impact can be significant. Median energy use per LLM query is around 0.34 Wh, rising to several Wh for long prompts, and large-scale or repeated use can accumulate"},{"citing_arxiv_id":"2603.26603","ref_index":77,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Sustainability Is Not Linear: Quantifying Performance, Energy, and Privacy Trade-offs in On-Device Intelligence","primary_cat":"cs.SE","submitted_at":"2026-03-27T17:00:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Empirical case study on a flagship Android device profiles energy, latency, and quality trade-offs across eight LLMs, revealing a quantization energy paradox and identifying mid-sized models as practical sweet spots.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2603.16951","ref_index":32,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Minimum-Action Learning: Energy-Constrained Symbolic Model Selection for Physical Law Identification from Noisy Data","primary_cat":"cs.LG","submitted_at":"2026-03-16T20:45:49+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"MAL recovers correct symbolic force laws like Kepler gravity from noisy data by minimizing trajectory reconstruction, sparsity, and energy violation, reaching 100% identification via energy criterion on benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}