arXiv:2506.17871v4 Announce Type: replace-cross
Abstract: Despite their impressive capabilities, aligned large language models (LLMs) often generate outputs that lack diversity. What drives this cons...
By Chenghao Yang, Sida Li, Ari Holtzman
Warning: This paper studies stereotypes and biases, and contains potentially disturbing examples, used for illustration purposes only. Our findings should not be interpreted as an argument against alignment.
arXiv:2609.23065v1 Announce Type: cross
Abstract: Chain-of-thought (CoT) can sound plausible yet be unfaithful to the model's underlying reasoning. Most prior work probes CoT faithfulness through inp...
By Qianli Wang, Yilong Wang, Dennis Wei, Jingyi Sun, Simon Ostermann, Pepa Atanasova, Nils Feldhus
arXiv:2606. 09850v1 Announce Type: new Abstract: Post-training alignment algorithms are predominantly evaluated as black boxes, obscuring how they reshape language models' internal computations.
By Aarush Sinha, Ishan Garg, Veeraraju Elluru, Arth Singh, Kushal Garg
The study investigates how post‑training of large autoregressive language models (ARMs) into masked diffusion models (MDMs) affects their internal computation. Across two 7B ARM‑MDM families and four diagnostic tasks, the authors find that MDMs retain much of the ARM’s high‑attribution pathways on prefix‑dominant tasks, but reorganize computation toward earlier layers on globally constrained tasks. Component‑level probes reveal that ARMs depend on sharply specialized components, whereas MDMs show weaker specialization and more diffuse output‑space alignment.
By Injin Kong, Hyoungjoon Lee, Yohan Jo
The paper extends mechanistic interpretability of large language models by modeling concepts as low‑dimensional non‑linear manifolds rather than linear subspaces. It introduces a concept‑based alignment (CBA) score to compare these manifolds across layers and models, revealing block structures in intermediate layers, a shift from syntax‑dominated to mixed syntactic‑semantic concepts, and training‑dependent multilingual sharing. The study also shows that alignment patterns differ across model families and training stages, with adjacent stages aligning more closely than distant ones.
By Tido Specht, Elias Benedict Krey, Nils Neukirch, Nils Strodthoff