IBM open-sources Steerability toolkit with steering pipelines and vLLM-Hook plugin
krvarshney · x · 2026-09-11
IBM's team announced the Steerability toolkit, aiming to bring structure to the scattered landscape of model steering methods. Highlights:
- Well-designed abstractions to organize the field's diverse steering techniques;
- Steering methods can be composed into chained "steering pipelines";
- Compatible with PEFT/TRL (training) and Inspect (evals);
- Ships with vLLM-Hook, a plugin for inspecting and steering model internals (attention patterns, heads, activations) on the vLLM engine, presented at ICML 2026.
Repo and docs are open source, with more features in progress.
More from Research
- K2 Horizon releases intermediate checkpoints, audits out 103 benchmark cheats — rohanpaul_ai · 2026-09-11
- 9th VISxAI workshop on AI explainability opens call at IEEE VIS 2026 in Boston — leland_mcinnes · 2026-09-11
- Researcher calls for perturbation-based multi-agent studies over one-off swarm observations — sebkrier · 2026-09-11
- Should researchers drop their marginal papers? A call for an RCT — RishiBommasani · 2026-09-11
- Science Advances paper introduces Media Bias Detector to measure publisher bias at scale — duncanjwatts · 2026-09-11
- OpenAI reportedly aims its new internal model at Riemann and P vs NP — zephyr_z9 · 2026-09-11