ContrastiveGuidance

steerability.algorithms.output_control.contrastive_guidance

args

control

ContrastiveGuidance

Bases: OutputControl

Contrastive mixing of next-token distributions as configuration.

ContrastiveGuidance is the generic over the distribution shape. It exposes the common logit source slot through flat Args: steer() resolves a parallel list of sources and mixes their log-probs with the base distribution as base_weight * log p_base + sum_i weights[i] * log p_source_i, with an optional plausibility mask. A method from the literature is an assignment of a config:

- DExperts: `sources=[expert_id, anti_expert_id], weights=[+α, -α]`.
- Contrastive decoding: `sources=[amateur_id], weights=[-1.0], alpha=0.1`.
- Proxy-tuning: `sources=[tuned_small_id, untuned_small_id], weights=[+1.0, -1.0]`.
- CFG / context-aware decoding: `sources=[{"kind": "prompt_variant", ...}], base_weight=gamma,
  weights=[-(gamma-1)]` (a `prompt_variant` source forwards the pipeline's own model; see the
  compatibility note on `PromptVariantSource`).

ContrastiveGuidance is a step-level control: it composes a ContrastiveMixtureProcessor into the decoding stack, so it works alongside other output controls and any decoding driver. Auxiliary sources must share the base vocabulary (enforced by AuxModelSource). supports_batching is False; generation runs one prompt at a time.

Reference:

• "DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts" Alisa Liu, Maarten Sap, Ximing Lu, Swabha Swayamdipta, Chandra Bhagavatula, Noah A. Smith, Yejin Choi https://arxiv.org/abs/2105.03023

• "Contrastive Decoding: Open-ended Text Generation as Optimization" Xiang Lisa Li, Ari Holtzman, Daniel Fried, Percy Liang, Jason Eisner, Tatsunori Hashimoto, Luke Zettlemoyer, Mike Lewis https://arxiv.org/abs/2210.15097

Source code in steerability/algorithms/output_control/contrastive_guidance/control.py

class ContrastiveGuidance(OutputControl):
    """Contrastive mixing of next-token distributions as configuration.

    `ContrastiveGuidance` is the generic over the distribution shape. It exposes the `common` logit
    source slot through flat `Args`: `steer()` resolves a parallel list of sources and mixes their
    log-probs with the base distribution as `base_weight * log p_base + sum_i weights[i] * log
    p_source_i`, with an optional plausibility mask. A method from the literature is an assignment of
    a config:

        - DExperts: `sources=[expert_id, anti_expert_id], weights=[+α, -α]`.
        - Contrastive decoding: `sources=[amateur_id], weights=[-1.0], alpha=0.1`.
        - Proxy-tuning: `sources=[tuned_small_id, untuned_small_id], weights=[+1.0, -1.0]`.
        - CFG / context-aware decoding: `sources=[{"kind": "prompt_variant", ...}], base_weight=gamma,
          weights=[-(gamma-1)]` (a `prompt_variant` source forwards the pipeline's own model; see the
          compatibility note on `PromptVariantSource`).

    `ContrastiveGuidance` is a step-level control: it composes a `ContrastiveMixtureProcessor` into
    the decoding stack, so it works alongside other output controls and any decoding driver.
    Auxiliary sources must share the base vocabulary (enforced by `AuxModelSource`).
    `supports_batching` is False; generation runs one prompt at a time.

    Reference:

    - "DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts"
      Alisa Liu, Maarten Sap, Ximing Lu, Swabha Swayamdipta, Chandra Bhagavatula, Noah A. Smith,
      Yejin Choi
      [https://arxiv.org/abs/2105.03023](https://arxiv.org/abs/2105.03023)

    - "Contrastive Decoding: Open-ended Text Generation as Optimization"
      Xiang Lisa Li, Ari Holtzman, Daniel Fried, Percy Liang, Jason Eisner, Tatsunori Hashimoto,
      Luke Zettlemoyer, Mike Lewis
      [https://arxiv.org/abs/2210.15097](https://arxiv.org/abs/2210.15097)
    """

    Args = ContrastiveGuidanceArgs

    # placeholders (filled by steer)
    tokenizer: PreTrainedTokenizerBase | None = None
    _sources: list | None = None

    def steer_access(self) -> ModelAccess:
        """`ModelAccess.MODULE`; sources may bind the live model (the prompt-variant source
        forwards it during decoding), which is retained past steer (the generate phase is
        in-process)."""
        return ModelAccess.MODULE

    def steer(
        self,
        model: PreTrainedModel,
        tokenizer: PreTrainedTokenizerBase | None = None,
        **__,
    ) -> PreTrainedModel:
        """Resolve each source spec (loading auxiliary models and enforcing shared vocabularies)."""
        self.tokenizer = tokenizer or getattr(model, "tokenizer", None)
        self._sources = [
            resolve_source(spec, model=model, tokenizer=self.tokenizer) for spec in self.sources
        ]
        self.same_model_forwards = any(
            getattr(source, "same_model_forwards", False) for source in self._sources
        )
        return model

    def get_logits_processors(self, input_ids, runtime_kwargs, **kwargs) -> list:
        """Return a fresh `ContrastiveMixtureProcessor` mixing the resolved sources."""
        if self._sources is None:
            raise RuntimeError("ContrastiveGuidance.steer() must run before generation (sources not loaded).")
        return [
            ContrastiveMixtureProcessor(
                list(zip(self._sources, self.weights)),
                base_weight=self.base_weight,
                alpha=self.alpha,
            )
        ]

    def cleanup(self) -> None:
        """Release every resolved source to free auxiliary-model memory."""
        for source in self._sources or []:
            source.cleanup()
        self._sources = None
        self.tokenizer = None
        gc.collect()
        if torch.cuda.is_available():
            torch.cuda.empty_cache()

RUNTIME_KWARGS_SCHEMA = [] class-attribute instance-attribute

args = self.Args.validate(*args, **kwargs) instance-attribute

enabled = True class-attribute instance-attribute

include_in_scoring = True class-attribute instance-attribute

same_model_forwards = False class-attribute instance-attribute

supports_batching = False class-attribute instance-attribute

tokenizer = None class-attribute instance-attribute

cleanup()

Release every resolved source to free auxiliary-model memory.

Source code in steerability/algorithms/output_control/contrastive_guidance/control.py

def cleanup(self) -> None:
    """Release every resolved source to free auxiliary-model memory."""
    for source in self._sources or []:
        source.cleanup()
    self._sources = None
    self.tokenizer = None
    gc.collect()
    if torch.cuda.is_available():
        torch.cuda.empty_cache()

clone_for_call(seed=None)

A configuration-preserving shallow clone for one generation call.

The clone shares steer-time artifacts (memories, steering vectors, attached tokenizers) with the original but has its own attribute namespace, so per-call attribute mutation on the clone never races another call using the original. When seed is given and the control defines reseed(seed), the clone's client-side RNG is re-seeded.

Parameters:

Name	Type	Description	Default
seed	int | None	Optional seed forwarded to the clone's reseed().	None

Returns:

Type	Description
BaseControl	The clone.

Source code in steerability/algorithms/core/base_control.py

def clone_for_call(self, seed: int | None = None) -> "BaseControl":
    """A configuration-preserving shallow clone for one generation call.

    The clone shares steer-time artifacts (memories, steering vectors, attached tokenizers)
    with the original but has its own attribute namespace, so per-call attribute mutation on
    the clone never races another call using the original. When `seed` is given and the
    control defines `reseed(seed)`, the clone's client-side RNG is re-seeded.

    Args:
        seed: Optional seed forwarded to the clone's `reseed()`.

    Returns:
        The clone.
    """
    clone = copy.copy(self)
    if seed is not None:
        reseed = getattr(clone, "reseed", None)
        if callable(reseed):
            reseed(seed)
    return clone

export_constraint(runtime_kwargs=None)

The control's declarative constrained-decoding source, or None.

A control whose per-step masking compiles from a declarative source returns a ConstraintSource; on a backend advertising Capability.GUIDED_DECODING the pipeline renders it onto the engine's native structured-output parameters in place of the control's live processor. The default returns None, which keeps the control on the live processor mechanism.

Parameters:

Name	Type	Description	Default
runtime_kwargs	dict | None	Per-call parameters supplied to generate().	None

Returns:

Type	Description
ConstraintSource | None	The constraint source, or None.

Source code in steerability/algorithms/output_control/base.py

def export_constraint(self, runtime_kwargs: dict | None = None) -> ConstraintSource | None:
    """The control's declarative constrained-decoding source, or None.

    A control whose per-step masking compiles from a declarative source returns a
    `ConstraintSource`; on a backend advertising `Capability.GUIDED_DECODING` the pipeline
    renders it onto the engine's native structured-output parameters in place of the
    control's live processor. The default returns None, which keeps the control on the live
    processor mechanism.

    Args:
        runtime_kwargs: Per-call parameters supplied to `generate()`.

    Returns:
        The constraint source, or None.
    """
    return None

export_generation_params(runtime_kwargs=None)

The control's sampling-expressible contribution, or None.

A control whose behavior is expressible as normalized generation parameters returns a mapping over a subset of stop_strings, stop_token_ids, max_new_tokens, and min_new_tokens; the pipeline merges it into the call's GenerationParams (stop rules union with the caller's; token bounds only tighten) and does not additionally collect the control's live processors and criteria for that call, so the control executes on every backend through the session's composed stop rules. The default returns None, which keeps the control on the live processor/criteria mechanism.

Parameters:

Name	Type	Description	Default
runtime_kwargs	dict | None	Per-call parameters supplied to generate().	None

Returns:

Type	Description
Mapping[str, Any] | None	The parameter contribution, or None.

Source code in steerability/algorithms/output_control/base.py

def export_generation_params(self, runtime_kwargs: dict | None = None) -> Mapping[str, Any] | None:
    """The control's sampling-expressible contribution, or None.

    A control whose behavior is expressible as normalized generation parameters returns a
    mapping over a subset of `stop_strings`, `stop_token_ids`, `max_new_tokens`, and
    `min_new_tokens`; the pipeline merges it into the call's `GenerationParams` (stop rules
    union with the caller's; token bounds only tighten) and does not additionally collect
    the control's live processors and criteria for that call, so the control executes on
    every backend through the session's composed stop rules. The default returns None, which
    keeps the control on the live processor/criteria mechanism.

    Args:
        runtime_kwargs: Per-call parameters supplied to `generate()`.

    Returns:
        The parameter contribution, or None.
    """
    return None

export_processor_spec(runtime_kwargs=None)

The control's engine-hosted processor form, or None.

A control whose per-step logit math is expressible in an engine's served processor vocabulary returns a ProcessorSpec; on a backend advertising Capability.PER_STEP_LOGIT_SPECS with the spec's kind, the pipeline submits it as a ProcessorSpecEntry in place of the control's live processor. The default returns None, which keeps the control on the live processor mechanism.

Parameters:

Name	Type	Description	Default
runtime_kwargs	dict | None	Per-call parameters supplied to generate().	None

Returns:

Type	Description
ProcessorSpec | None	The processor spec, or None.

Source code in steerability/algorithms/output_control/base.py

def export_processor_spec(self, runtime_kwargs: dict | None = None) -> ProcessorSpec | None:
    """The control's engine-hosted processor form, or None.

    A control whose per-step logit math is expressible in an engine's served processor
    vocabulary returns a `ProcessorSpec`; on a backend advertising
    `Capability.PER_STEP_LOGIT_SPECS` with the spec's kind, the pipeline submits it as a
    `ProcessorSpecEntry` in place of the control's live processor. The default returns
    None, which keeps the control on the live processor mechanism.

    Args:
        runtime_kwargs: Per-call parameters supplied to `generate()`.

    Returns:
        The processor spec, or None.
    """
    return None

export_state()

Steer-time state to persist when freezing, keyed by logical name.

Values are typed artifacts (SteeringVector, Probe, ProbeSet, a Memory implementation, CheckpointArtifact, LoRAArtifact, torch.Tensor, or a Path to an on-disk product). Must be called after steer(). The default returns an empty mapping (nothing to persist).

Returns:

Type	Description
dict[str, Any]	Mapping from logical name to artifact value.

Source code in steerability/algorithms/core/base_control.py

def export_state(self) -> dict[str, Any]:
    """Steer-time state to persist when freezing, keyed by logical name.

    Values are typed artifacts (`SteeringVector`, `Probe`, `ProbeSet`, a `Memory`
    implementation, `CheckpointArtifact`, `LoRAArtifact`, `torch.Tensor`, or a `Path` to an
    on-disk product). Must be called after `steer()`. The default returns an empty mapping
    (nothing to persist).

    Returns:
        Mapping from logical name to artifact value.
    """
    return {}

fit_identity()

The object whose canonical form defines this control's fit digest, or None.

The digest detects recipe edits that invalidate frozen artifacts, so the returned object should cover exactly the inputs a re-fit would consume (training data, fit specs) and exclude inert application parameters (strengths, multipliers). The default returns None (no fits).

Returns:

Type	Description
Any | None	The fit-identity object, or None.

Source code in steerability/algorithms/core/base_control.py

def fit_identity(self) -> Any | None:
    """The object whose canonical form defines this control's fit digest, or None.

    The digest detects recipe edits that invalidate frozen artifacts, so the returned
    object should cover exactly the inputs a re-fit would consume (training data, fit
    specs) and exclude inert application parameters (strengths, multipliers). The default
    returns None (no fits).

    Returns:
        The fit-identity object, or None.
    """
    return None

frozen_form(state)

The (registry method key, constructor kwargs) of this control's frozen form.

state is the output of export_state(); artifact values inside the returned kwargs are replaced by store references at encode time. When steer_fits() is empty and state is empty, the recipe is the frozen form and the control's own key and recipe args are returned unchanged. A control whose freeze expands into several entries may return a list of (method key, kwargs) pairs instead.

Parameters:

Name	Type	Description	Default
state	dict[str, Any]	The output of export_state().	required

Returns:

Type	Description
tuple[str, dict[str, Any]]	The frozen method key and its constructor kwargs.

Raises:

Type	Description
NotFreezableError	If the control produces fits or state but declares no frozen form.

Source code in steerability/algorithms/core/base_control.py

def frozen_form(self, state: dict[str, Any]) -> tuple[str, dict[str, Any]]:
    """The `(registry method key, constructor kwargs)` of this control's frozen form.

    `state` is the output of `export_state()`; artifact values inside the returned kwargs
    are replaced by store references at encode time. When `steer_fits()` is empty and
    `state` is empty, the recipe is the frozen form and the control's own key and recipe
    args are returned unchanged. A control whose freeze expands into several entries may
    return a list of `(method key, kwargs)` pairs instead.

    Args:
        state: The output of `export_state()`.

    Returns:
        The frozen method key and its constructor kwargs.

    Raises:
        NotFreezableError: If the control produces fits or state but declares no frozen
            form.
    """
    if not state and not self.steer_fits():
        from steerability.algorithms.core.registry import method_key_for

        args = getattr(self, "args", None)
        kwargs = {field.name: getattr(args, field.name) for field in fields(args) if field.init} \
            if args is not None else {}
        return method_key_for(type(self)), kwargs
    raise NotFreezableError(
        f"{type(self).__name__} produces steer-time state but declares no frozen form; "
        "override frozen_form() (and export_state()) to support freezing."
    )

get_logits_processors(input_ids, runtime_kwargs, **kwargs)

Return a fresh ContrastiveMixtureProcessor mixing the resolved sources.

Source code in steerability/algorithms/output_control/contrastive_guidance/control.py

def get_logits_processors(self, input_ids, runtime_kwargs, **kwargs) -> list:
    """Return a fresh `ContrastiveMixtureProcessor` mixing the resolved sources."""
    if self._sources is None:
        raise RuntimeError("ContrastiveGuidance.steer() must run before generation (sources not loaded).")
    return [
        ContrastiveMixtureProcessor(
            list(zip(self._sources, self.weights)),
            base_weight=self.base_weight,
            alpha=self.alpha,
        )
    ]

get_stopping_criteria(input_ids, runtime_kwargs, **kwargs)

The control's stopping criteria.

Not applied during scoring (there is no loop to stop). Same call convention as get_logits_processors.

Parameters:

Name	Type	Description	Default
input_ids	Tensor	The steered prompt token ids [batch, seq_len].	required
runtime_kwargs	dict | None	Per-call parameters supplied to generate().	required

Returns:

Type	Description
list	A list of HF StoppingCriteria-style objects.

Source code in steerability/algorithms/output_control/base.py

def get_stopping_criteria(self, input_ids: torch.Tensor, runtime_kwargs: dict | None, **kwargs) -> list:
    """The control's stopping criteria.

    Not applied during scoring (there is no loop to stop). Same call convention as
    `get_logits_processors`.

    Args:
        input_ids: The steered prompt token ids `[batch, seq_len]`.
        runtime_kwargs: Per-call parameters supplied to `generate()`.

    Returns:
        A list of HF `StoppingCriteria`-style objects.
    """
    return []

requirements()

Backend requirements computed from this instance's configuration, per phase.

The default requires Capability.IN_PROCESS_TORCH at generate and, when include_in_scoring is True, at score as well, since remote prompt-logprob computation applies neither live processors nor engine-registered sampling processors to prefill logits. Setting include_in_scoring=False removes the score-phase requirement.

Returns:

Type	Description
Requirements	The control's phase-keyed requirements.

Source code in steerability/algorithms/output_control/base.py

def requirements(self) -> Requirements:
    """Backend requirements computed from this instance's configuration, per phase.

    The default requires `Capability.IN_PROCESS_TORCH` at generate and, when
    `include_in_scoring` is True, at score as well, since remote prompt-logprob computation
    applies neither live processors nor engine-registered sampling processors to prefill
    logits. Setting `include_in_scoring=False` removes the score-phase requirement.

    Returns:
        The control's phase-keyed requirements.
    """
    score = needs(Capability.IN_PROCESS_TORCH) if self.include_in_scoring else ()
    return Requirements(generate=needs(Capability.IN_PROCESS_TORCH), score=score)

steer(model, tokenizer=None, **__)

Resolve each source spec (loading auxiliary models and enforcing shared vocabularies).

Source code in steerability/algorithms/output_control/contrastive_guidance/control.py

def steer(
    self,
    model: PreTrainedModel,
    tokenizer: PreTrainedTokenizerBase | None = None,
    **__,
) -> PreTrainedModel:
    """Resolve each source spec (loading auxiliary models and enforcing shared vocabularies)."""
    self.tokenizer = tokenizer or getattr(model, "tokenizer", None)
    self._sources = [
        resolve_source(spec, model=model, tokenizer=self.tokenizer) for spec in self.sources
    ]
    self.same_model_forwards = any(
        getattr(source, "same_model_forwards", False) for source in self._sources
    )
    return model

steer_access()

ModelAccess.MODULE; sources may bind the live model (the prompt-variant source forwards it during decoding), which is retained past steer (the generate phase is in-process).

Source code in steerability/algorithms/output_control/contrastive_guidance/control.py

def steer_access(self) -> ModelAccess:
    """`ModelAccess.MODULE`; sources may bind the live model (the prompt-variant source
    forwards it during decoding), which is retained past steer (the generate phase is
    in-process)."""
    return ModelAccess.MODULE

steer_fits()

The fit artifacts this instance's steer step will produce, for the steer plan.

Each entry is (artifact, artifact_class), where artifact is the fit source or recipe class name and artifact_class is "direction" or "calibrated". The default is an empty tuple (no fits).

Returns:

Type	Description
tuple[tuple[str, str], ...]	The declared fit artifacts, in declaration order.

Source code in steerability/algorithms/core/base_control.py

def steer_fits(self) -> tuple[tuple[str, str], ...]:
    """The fit artifacts this instance's steer step will produce, for the steer plan.

    Each entry is `(artifact, artifact_class)`, where `artifact` is the fit source or
    recipe class name and `artifact_class` is `"direction"` or `"calibrated"`. The default
    is an empty tuple (no fits).

    Returns:
        The declared fit artifacts, in declaration order.
    """
    return ()