Cover Image

When Shallow Wins: Silent Failures and the Depth-Accuracy Paradox in Latent Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Sahoo, Subramanyam, Chadha, Aman, Jain, Vinija, Chaudhary, Divya
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Artificial Intelligence Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

The Reasoning Trap -- Logical Reasoning as a Mechanistic Pathway to Situational Awareness
by: Sahoo, Subramanyam, et al.
Published: (2026)

SAHOO: Safeguarded Alignment for High-Order Optimization Objectives in Recursive Self-Improvement
by: Sahoo, Subramanyam, et al.
Published: (2026)

I Can't Believe It's Not Robust: Catastrophic Collapse of Safety Classifiers under Embedding Drift
by: Sahoo, Subramanyam, et al.
Published: (2026)

Are Small Language Models Ready to Compete with Large Language Models for Practical Applications?
by: Sinha, Neelabh, et al.
Published: (2024)

Position: The Complexity of Perfect AI Alignment -- Formalizing the RLHF Trilemma
by: Sahoo, Subramanyam, et al.
Published: (2025)

Guiding Vision-Language Model Selection for Visual Question-Answering Across Tasks, Domains, and Knowledge Types
by: Sinha, Neelabh, et al.
Published: (2024)

Decoding the Diversity: A Review of the Indic AI Research Landscape
by: KJ, Sankalp, et al.
Published: (2024)

MENTIS: What Belief Changes Under Alignment? Measuring Multi-Scale Latent Torsion in Language Models
by: Saha, Partha Pratim, et al.
Published: (2026)

How Culturally Aware are Vision-Language Models?
by: Burda-Lassen, Olena, et al.
Published: (2024)

Parameter Efficient Fine Tuning: A Comprehensive Analysis Across Applications
by: Balne, Charith Chandra Sai, et al.
Published: (2024)

Catch Me If You Can: How Smaller Reasoning Models Pretend to Reason with Mathematical Fidelity
by: Sahoo, Subramanyam, et al.
Published: (2025)

SPINAL -- Scaling-law and Preference Integration in Neural Alignment Layers
by: Das, Arion, et al.
Published: (2026)

Calibration Collapse Under Sycophancy Fine-Tuning: How Reward Hacking Breaks Uncertainty Quantification in LLMs
by: Sahoo, Subramanyam
Published: (2026)

Dial E for Ethical Enforcement: institutional VETO power as a governance primitive
by: Sahoo, Subramanyam, et al.
Published: (2026)

PermaFrost-Attack: Stealth Pretraining Seeding(SPS) for planting Logic Landmines During LLM Training
by: Kumar, Harsh, et al.
Published: (2026)

A Comprehensive Survey of Hallucination in Large Language, Image, Video and Audio Foundation Models
by: Sahoo, Pranab, et al.
Published: (2024)

Overview of Factify5WQA: Fact Verification through 5W Question-Answering
by: Suresh, Suryavardan, et al.
Published: (2024)

AlignGuard-LoRA: Alignment-Preserving Fine-Tuning via Fisher-Guided Decomposition and Riemannian-Geodesic Collision Regularization
by: Das, Amitava, et al.
Published: (2025)

Born With a Silver Spoon? Investigating Socioeconomic Bias in Large Language Models
by: Singh, Smriti, et al.
Published: (2024)

Assessing LLM Reliability on Temporally Recent Open-Domain Questions
by: Krishnappa, Pushwitha, et al.
Published: (2026)

Neural FOXP2 -- Language Specific Neuron Steering for Targeted Language Improvement in LLMs
by: Saha, Anusa, et al.
Published: (2026)

A Comprehensive Survey of Accelerated Generation Techniques in Large Language Models
by: Khoshnoodi, Mahsa, et al.
Published: (2024)

Multilingual State Space Models for Structured Question Answering in Indic Languages
by: Vats, Arpita, et al.
Published: (2025)

ECLIPTICA -- A Framework for Switchable LLM Alignment via CITA - Contrastive Instruction-Tuned Alignment
by: Wanaskar, Kapil, et al.
Published: (2026)

A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications
by: Sahoo, Pranab, et al.
Published: (2024)

Can Large Language Models Infer Causal Relationships from Real-World Text?
by: Saklad, Ryan, et al.
Published: (2025)

TRACEALIGN -- Tracing the Drift: Attributing Alignment Failures to Training-Time Belief Sources in LLMs
by: Das, Amitava, et al.
Published: (2025)

Depth-Recurrent Attention Mixtures: Giving Latent Reasoning the Attention it Deserves
by: Knupp, Jonas, et al.
Published: (2026)

Large Language and Reasoning Models are Shallow Disjunctive Reasoners
by: Khalid, Irtaza, et al.
Published: (2025)

DPO Kernels: A Semantically-Aware, Kernel-Enhanced, and Divergence-Rich Paradigm for Direct Preference Optimization
by: Das, Amitava, et al.
Published: (2025)

Investigating Annotator Bias in Large Language Models for Hate Speech Detection
by: Das, Amit, et al.
Published: (2024)

When Two LLMs Debate, Both Think They'll Win
by: Prasad, Pradyumna Shyama, et al.
Published: (2025)

Exploring the Frontier of Vision-Language Models: A Survey of Current Methodologies and Future Directions
by: Ghosh, Akash, et al.
Published: (2024)

When Greedy Wins: Emergent Exploitation Bias in Meta-Bandit LLM Training
by: Chen, Sanxing, et al.
Published: (2025)

MAAT: Multi-phase Adapter-Aware Targeted Unlearning
by: Yagnik, Suryash, et al.
Published: (2026)

When Domains Interact: Asymmetric and Order-Sensitive Cross-Domain Effects in Reinforcement Learning for Reasoning
by: Yang, Wang, et al.
Published: (2026)

Certifying Knowledge Comprehension in LLMs
by: Chaudhary, Isha, et al.
Published: (2024)

SKETCH: Structured Knowledge Enhanced Text Comprehension for Holistic Retrieval
by: Mahalingam, Aakash, et al.
Published: (2024)

Mini-Giants: "Small" Language Models and Open Source Win-Win
by: Zhou, Zhengping, et al.
Published: (2023)

AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning dataset
by: Moshkov, Ivan, et al.
Published: (2025)