7 papers
Position: Don't Just "Fix it in Post": A Science of AI Must Study Training Dynamics
Stella Biderman, Mohammad Aflah Khan, Niloofar Mireshghallah +3
What would it mean to have a scientific understanding of AI? Models are not static objects: they are snapshots of time-evolving processes shaped by data, objectives, architectures,…
Fine-tuning vs. In-context Learning in Large Language Models: A Formal Language Learning Perspective
Bishwamittra Ghosh, Soumi Das, Till Speicher +5
Large language models (LLMs) operate in two fundamental learning modes - fine-tuning (FT) and in-context learning (ICL) - raising key questions about which mode yields greater lang…
Fractional Rotation, Full Potential? Investigating Performance and Convergence of Partial RoPE
Mohammad Aflah Khan, Krishna P. Gummadi, Manish Gupta +1
Rotary Positional Embedding (RoPE) is a common choice in transformer architectures for encoding relative positional information. Although earlier work has examined omitting RoPE in…
Rote Learning Considered Useful: Generalizing over Memorized Data in LLMs
Qinyuan Wu, Soumi Das, Mahsa Amani +4
Rote learning is a memorization technique based on repetition. Many researchers argue that rote learning hinders generalization because it encourages verbatim memorization rather t…
In Agents We Trust, but Who Do Agents Trust? Latent Source Preferences Steer LLM Generations
Mohammad Aflah Khan, Mahsa Amani, Soumi Das +5
Agents based on Large Language Models (LLMs) are increasingly being deployed as interfaces to information on online platforms. These agents filter, prioritize, and synthesize infor…
Rethinking Memorization Measures and their Implications in Large Language Models
Bishwamittra Ghosh, Soumi Das, Qinyuan Wu +4
Concerned with privacy threats, memorization in LLMs is often seen as undesirable, specifically for learning. In this paper, we study whether memorization can be avoided when optim…