Get started
Author

Subramanyam Sahoo

When AI Benchmarks Plateau: A Systematic Study of Benchmark Saturation
2026 · Akhtar, Mubashara, Reuel, Anka, et al. · arXiv (Cornell University) · 3
PDF
Boardwalk Empire: How Generative AI is Revolutionizing Economic Paradigms
2024 · Sahoo, Subramanyam, Dutta, Kamlesh · arXiv (Cornell University) · 3
PDF
From Knowledge to Action: Outcomes of the 2025 Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry
2026 · Roy, Aritra, Shen, Kevin, et al. · arXiv (Cornell University) · 1
PDF
DUNE: Decoding Unified Naive Bayes Explainability through Gaussian methods for a Heart Disease Diagnostic
2024 · Subramanyam Sahoo, Kamlesh Dutta · International Conference on Computing, Communication and Networking Technologies (ICCCNT) · 1
The Curious Case of "NETFLIX": Predictive Analysis and Preferences for Indian Users
2024 · Subramanyam Sahoo, Kamlesh Dutta · SSRN Electronic Journal · 1
Evaluation of Standard Classifiers for Protein Subcellular Localization
2020 · Debasis Mohapatra, Sayoni Das, et al. · 1
Better Call Reward: Reward Hacking as Strategic Abstention in Legal Reasoning Models
2026 · Sahoo, Subramanyam, Shenk, Justin · arXiv (Cornell University)
PDF
Pessimism's Paradox: Conservative Offline Training Amplifies Reward Hacking During Online Adaptation in Reasoning Models
2026 · Sahoo, Subramanyam, Chadha, Aman, et al. · arXiv (Cornell University)
PDF
Creative Collision: Directorial Persona Steering and Competition in Large Language Models
2026 · Sahoo, Subramanyam, Shenk, Justin · arXiv (Cornell University)
PDF
Evaluation Cards: An Interpretive Layer for AI Evaluation Reporting
2026 · Ghosh, Avijit, Reuel, Anka, et al. · arXiv (Cornell University)
PDF
Calibration of Structured Ignorance Certificates for Diagnosing Unknown Unknowns in Reasoning Models
2026 · Sahoo, Subramanyam · arXiv (Cornell University)
PDF
Memetic Capture: A Pluralistic Policy Framework for Governing AI-Driven Cultural Disempowerment
2026 · Sahoo, Subramanyam · arXiv (Cornell University)
PDF
Linear Probes Detect Task Format, Not Reasoning Mode in Language Model Hidden States
2026 · Sahoo, Subramanyam, Jain, Vinija, et al. · arXiv (Cornell University)
PDF
Calibration Collapse Under Sycophancy Fine-Tuning: How Reward Hacking Breaks Uncertainty Quantification in LLMs
2026 · Sahoo, Subramanyam · arXiv (Cornell University)
PDF
The Reasoning Trap -- Logical Reasoning as a Mechanistic Pathway to Situational Awareness
2026 · Sahoo, Subramanyam, Chadha, Aman, et al. · arXiv (Cornell University)
PDF
SAHOO: Safeguarded Alignment for High-Order Optimization Objectives in Recursive Self-Improvement
2026 · Sahoo, Subramanyam, Chadha, Aman, et al. · arXiv (Cornell University)
PDF
Policy myopia as a mechanism of gradual disempowerment in Post-AGI governance, Circa 2049
2026 · Sahoo, Subramanyam · arXiv (Cornell University)
PDF
When Shallow Wins: Silent Failures and the Depth-Accuracy Paradox in Latent Reasoning
2026 · Sahoo, Subramanyam, Chadha, Aman, et al. · arXiv (Cornell University)
PDF
The Controllability Trap: A Governance Framework for Military AI Agents
2026 · Sahoo, Subramanyam · arXiv (Cornell University)
PDF
I Can't Believe It's Not Robust: Catastrophic Collapse of Safety Classifiers under Embedding Drift
2026 · Sahoo, Subramanyam, Jain, Vinija, et al. · arXiv (Cornell University)
PDF
Dial E for Ethical Enforcement: institutional VETO power as a governance primitive
2026 · Sahoo, Subramanyam, Jain, Vinija, et al. · arXiv (Cornell University)
PDF
Position: The Reasoning Trap -- Logical Reasoning as a Mechanistic Pathway to Situational Awareness
2026 · Subramanyam Sahoo, Chadha, Aman, et al.
Linear Probes Detect Task Format, Not Reasoning Mode in Language Model Hidden States
2026 · Subramanyam Sahoo, Vinija Jain, et al. · Workshop on Trustworthy NLP (TrustNLP)
PDF
The Green Mile: A Multi-Layered Quest to Reveal, Measure, and Slash Carbon Emissions in AI Training & Inference
2025 · Subramanyam Sahoo, Rajesh Kesavalalji, et al. · Indian Conference on Computer Vision, Graphics and Image Processing
The Double Life of Code World Models: Provably Unmasking Malicious Behavior Through Execution Traces
2025 · Sahoo, Subramanyam · arXiv (Cornell University)
PDF
Works count: 30
Citation count: 10
H-index: 2
I10-index:
Year
Topic
Type