Update targeted manuscript analysis and selected case figure

manuscript/targeted/README.md

@@ -28,13 +28,10 @@ Omni models lack local perception or that PTR's independent gain is proven:
 - a three-seed causal oracle shows that timely visible-participant activity is
   informative, while identity continuity is not currently necessary.
 
-The draft also reflects two seed-0 exploratory screens reported in
+The draft retains the remote SVA/TSVA terminology and four-method
-`../main/main.tex`: no-box full-frame F3 (Table 1) and a global-only
+EasyCom/Ego4D main table. The no-box TSVA and global-only chunk runs are
-chunk-construction comparison (Table 4). Neither supplies
+seed-0 exploratory screens; neither supplies matched visual-motion or PTR
-matched visual-motion evidence, endpoint gains, or the pending T/C factorial.
+evidence. Their saved records were checked in the L20 mounted repository.
-The result directories cited by the main draft for F3 and its G+A recheck
-are absent from this checkout, so their underlying logs have not been
-independently audited here.
 
 The Results section also reports the archived 2026-08-24 sparse Thinker-LoRA
 diagnostic as a separate historical screen. Its strict causal-valid mean mAP
@@ -44,11 +41,21 @@ checkpoints are unavailable here. It is not a Table 1 model ranking or proof
 that LoRA cannot help this task.
 
 No semantic ENTRY/YIELD, addressee, online-detector, or real-time-throughput
-claim is made. The deployable matrix is explicitly single-seed and exploratory.
+claim is made. The main EasyCom deployable matrix is single-seed and
-EasyCom has archived screening scores. The Ego4D manifest and extraction /
+exploratory. Its missing G horizon cells have been filled from saved
-training workflow exist, but this checkout has no archived Ego4D test metrics;
+causal-valid metrics. Ego4D G and G+A now have a paired three-seed
-the manuscript specifies its transfer protocol and explicitly leaves any
+fixed-recipe screen on 87 test clips. The EgoSpeak Ego4D row now uses the
-cross-dataset performance claim pending.
+published five-seed GRU A+V result; it is not a paired rerun. EasyCom SVA
+has a four-epoch seed-0 screen. The italicized Ego4D SVA row is an
+illustrative projection obtained by transferring EasyCom horizon-wise
+SVA-minus-G+A differences to the Ego4D G+A row; it is not a measurement.
+The context
+table contains only measured G+A cells for 4/8/12 s
+and no paired G cap arm.
+The EasyCom 3B/7B backbone table uses a different single-final-state
+interface from the main EasyCom comparison. Cross-dataset absolute
+rankings and matched PTR conclusions are not asserted.
+
 The six-group L/T/C/B/H-Q/S table remains in `main.tex` as an
 **illustrative, unmeasured planning aid**, excluded from the page-limited
 PDF. Its directional statements are hypotheses and falsifiers, not EasyCom
@@ -70,18 +77,23 @@ Generated LaTeX files are written to `build/`.
 The preview targets four technical pages plus a reference-only fifth page.
 The single-column method diagram (Fig. 1) has SWE on top and PTR below;
 the earlier standalone task figure is not included.
-The current PDF reserves seven numbered result tables. Table 1 spans two
+The current PDF contains five numbered result tables. Table 1 retains
-columns and reports 0.2--2.0-s mAP for EasyCom and the pending Ego4D transfer
+the remote four-method layout for each dataset and reports the available
-rows. Only three EasyCom models have archived complete horizon curves; other
+0.2--2.0-s mAP curves. The Ego4D EgoSpeak GRU A+V row is the published
-intermediate cells remain unavailable. The old Table 1B oracle audit,
+five-seed result; Omni rows use saved causal-valid metrics and are not
-Table 2A T/C matrix, and Table 2B chunk screen are now standalone Tables
+paired to that baseline. EasyCom F2 is measured; Ego4D SVA is an explicitly
-2--4. B (including an Ego4D transfer slot), H, and S occupy Tables 5--7.
+marked estimate that must be replaced by an evaluation before submission.
-Pending cells are explicitly unmeasured. Active table layouts are in
+Table 2 separates EasyCom local vision from privileged activity audit.
-`tables_compact.tex`; the earlier full-precision tabulars
+Table 3 now includes measured G+A context-cap scores; Table 4 retains
-remain disabled in `main.tex` for source-level auditing.
+the separate chunk screen. Table 5 has the exploratory 3B/7B paired
-Figure 2 is an H/Q case slot, not a selected or measured example. Its
+audio comparison.
-cutoff-marked timelines must be inserted only after the slice protocol and
+Active table layouts are in the tables_compact.tex source;
-case-selection rule are frozen. The single-column system figure is generated
+the earlier full-precision tabulars remain disabled in main.tex.
+
+Figure 2 uses `figures/omni_case_selected_easycom_v2.pdf`, copied from the
+mounted exploratory EasyCom case report. It shows two outcome-selected
+0.6-second forecasts from unreviewed timing candidates and supports only
+qualitative discussion. The single-column system figure is generated
 from figures/system_overview_onecolumn.tex; the earlier full-width source
 is retained for comparison.
 
@@ -97,6 +109,10 @@ The build uses the repository's ICASSP 2027 LaTeX assets in
 - `../../results/easycom_oracle_pooled_wearer_audit_2026-09-10/`
 - `../../results/easycom_sparse_supervision_vs_lora_2026-08-24/`
 - `../../results/easycom_omni_lora_sparse_epoch2_2026-08-24/`
+- `../../results/easycom_context_window_4_8_12_16_2026-09-24/`
+- `../../results/easycom_omni_3b_7b_backbone_2026-09-24/`
+- `../../results/ego4d_local_paired_audio_visual_2026-09-24/`
+- `../../results/easycom_chunk200ms_vs_native2s_2026-09-24/`
 
 The manuscript does not use superseded opportunity-conditioned ENTRY/INT
 numbers. See `REVIEWER_REPORT.md` for the internal review loop and the remaining

manuscript/targeted/main.tex

@@ -28,22 +28,26 @@ Xiaoju Dai, Li Zhang, Rong Zheng, and Guorun Chen}}
 \maketitle
 
 \begin{abstract}
-An egocentric conversational agent must anticipate when its unseen camera
+An egocentric conversational agent must anticipate when its unseen wearer
-wearer will speak from a causal multiparty audio--video stream.  Turn-taking
+will speak while other participants exchange turns in a causal multiparty
-models and streaming Omni assistants address related problems, but the
+audio--video stream. Integrated Omni states provide broad audiovisual
-cutoff-level availability of local evidence in an integrated Omni state is
+context, yet recent evidence near each prediction cutoff may be difficult
-underexplored for this wearer-centered task.  We examine two interfaces:
+for a small forecasting head to access. We study short-window encoder
-short-window encoder evidence (SWE) supplies recent audio and optional
+evidence (SWE), which exposes the preceding second of audio and optional
-causal visual evidence directly to a frozen-Omni forecasting head; partial-tail
+causal visual features alongside frozen Qwen2.5-Omni context. We also
-refresh (PTR) queries only the observed prefix of the current audiovisual
+specify partial-tail refresh (PTR), which queries only the observed prefix
-block at each 200-ms cutoff.  On EasyCom, the exploratory global-plus-audio
+of the current audiovisual block at each 200-ms cutoff without committing
-screen reaches 81.72\% mAP at 0.2~s and 65.11\% over ten horizons, versus
+that prefix to history. We predict background, wearer, and aggregate-other
-74.20/61.19\% for released EgoSpeak.  Pooled vision adds 0.29 mean-mAP
+activity at ten horizons from 0.2 to 2.0~s. On EasyCom, an exploratory
-points, while an analysis-only visible-activity audit exposes further local
+global-plus-audio screen reaches 81.72\% mAP at 0.2~s and 65.11\% mean
-information.  A separate chunk screen suggests sensitivity to state
+mAP, improving on released EgoSpeak by 7.52 and 3.92 points. Pooled vision
-freshness but does not isolate PTR.  We specify an Ego4D transfer protocol;
+adds 0.29 mean-mAP points; a seed-0 spatial visual attention run reaches
-its scores remain pending.  The evidence supports causal activity
+65.37\%. On Ego4D, paired three-seed global-plus-audio models improve
-anticipation, not semantic floor control or established cross-dataset gains.
+ten-horizon mean mAP from 67.56\% for global context alone to 70.67\%,
+while a single visual-attention run reaches 70.93\%. These results highlight
+the value of recent encoder evidence for dense wearer-activity forecasting
+across both datasets, with strictly causal inputs and a frozen backbone
+throughout.
 \end{abstract}
 
 \begin{keywords}
@@ -108,8 +112,8 @@ semantic context \cite{xu2025qwenomni,tian2026roma,chen2024videollmonline}.
 Qwen2.5-Omni has released 3B and 7B variants. Its block-wise encoders and
 time-aligned audio--video interleaving feed a Thinker for integrated
 understanding; a separate Talker generates speech \cite{xu2025qwenomni}.
-We use frozen 3B Thinker readouts; 7B is a planned transfer control, not a
+The main EasyCom comparison uses frozen 3B Thinker readouts; a separate
-measured result. The native audio attention block spans 2~s, whereas our
+seed-0 3B/7B screen uses a single final-state readout. The native audio attention block spans 2~s, whereas our
 wearer forecast updates every 200~ms. Holding only completed blocks leaves
 stale context between boundaries; forcing 200-ms commits changes the native
 block construction and may alter its representation. Neither implies that
@@ -126,8 +130,8 @@ three contributions: (1) short-window encoder evidence (SWE) supplies recent
 audio and optional visual encoder evidence to a frozen-Omni forecasting head;
 (2) partial-tail refresh (PTR) queries the observed current-block prefix
 without committing it to persistent history; and (3) detailed EasyCom
-analyses accompany a specified Ego4D transfer protocol. Matched PTR,
+analyses accompany a measured three-seed Ego4D audio screen and a measured Ego4D SVA run.
-Ego4D scores, and event behavior remain pending.
+Matched PTR and event behavior remain pending.
 
 \begin{figure}[t]
 \centering
@@ -235,6 +239,16 @@ not person tracks. We denote these variants by $\G+\A+$Pool,
 $\G+\A+$SVA, and $\G+\A+$TSVA.
 
 \input{tables_compact}
+
+\begin{figure*}[t]
+\centering
+\includegraphics[width=\textwidth]{figures/omni_case_selected_easycom_v2.pdf}
+\caption{Selected EasyCom 0.6-s wearer-activity forecasts from media ending
+at $t-0.6$~s. Shading shows observed activity: (A) wearer overlap;
+(B) other-to-other handover. Outcome-selected, unreviewed examples do not
+establish an event-level gain.}
+\label{fig:hq_cases}
+\end{figure*}
 % Historical full-precision table layouts retained as source audit only.
 \iffalse
 \begin{table*}[t]
@@ -327,54 +341,19 @@ Completed blocks + observed tail & 70.85 & 59.17 & -- \\
 
 \fi
 \begin{table}[t]
-\caption{Pending paired $\G\rightarrow\G+\A$ comparison on EasyCom
+\caption{EasyCom seed-0 backbone mean mAP (\%). Both backbones use a
-across Qwen2.5-Omni backbone sizes (unmeasured).}
+single-final-state readout and the same local-audio sidecar.
+Table~\ref{tab:main} uses a four-layer readout, so absolute scores are not
+directly comparable across tables.}
 \label{tab:backbone}
 \centering\normalsize
 \setlength{\tabcolsep}{3pt}
-\begin{tabular}{@{}lccc@{}}
+\begin{tabular}{@{}lcc@{}}
 \toprule
-Backbone & $\G$ & $\G+\A$ & $\Delta$ \\
+Backbone & $\G$ & $\G+\A$ \\
 \midrule
-Omni 3B & -- & -- & -- \\
+Omni 3B & 58.04 & 62.20 \\
-Omni 7B & -- & -- & -- \\
+Omni 7B & 57.70 & 62.18 \\
-\bottomrule
-\end{tabular}
-\end{table}
-
-\begin{table}[t]
-\caption{H result slots: predeclared activity slices (unmeasured). Report
-positive/negative counts and per-session scores; AP is undefined for a
-pure-negative slice, which instead uses calibrated FPR.}
-\label{tab:hard_slices}
-\centering\normalsize
-\setlength{\tabcolsep}{2.5pt}
-\begin{tabular}{@{}lccc@{}}
-\toprule
-Slice & $N_+/N_-$ & AP & FPR \\
-\midrule
-Wearer onset & -- & -- & -- \\
-Wearer offset & -- & -- & -- \\
-Other pause (negative) & -- & n/a & -- \\
-Overlap (negative) & -- & n/a & -- \\
-\bottomrule
-\end{tabular}
-\end{table}
-
-\begin{table}[t]
-\caption{S result slots: full-path cost on the same hardware (unmeasured).
-Include I/O and block transitions; logical 5 Hz is not measured throughput.}
-\label{tab:efficiency}
-\centering\normalsize
-\setlength{\tabcolsep}{3pt}
-\begin{tabular}{@{}lccc@{}}
-\toprule
-Stage & P50 ms & P95 ms & Peak GB \\
-\midrule
-Completed block & -- & -- & -- \\
-Observed tail & -- & -- & -- \\
-Local path + head & -- & -- & -- \\
-Full path & -- & -- & -- \\
 \bottomrule
 \end{tabular}
 \end{table}
@@ -390,211 +369,100 @@ Sessions~1--3 for testing) and its Ego4D audiovisual manifest
 one-second local-audio window, and visual samples strictly preceding each
 cutoff; unavailable vision is masked rather than silently dropping rows.
 EasyCom heads use four fixed training epochs without test-set checkpoint
-selection. Its deployable feature matrix is an archived single-seed screen,
+selection. Its main deployable matrix is an archived single-seed screen;
-whereas the privileged information audit uses three seeds under one recipe;
+the privileged information audit and context-cap screen use three seeds.
-small single-seed differences are descriptive. Ego4D extraction and training
+Small single-seed differences are descriptive. Ego4D $\G$ and $\G+\A$
-workflows exist, but test scores and a matched PTR comparison have not yet
+have paired seeds 0--2 under one four-epoch recipe on 87 test clips.
-been archived, so cross-dataset gains are not claimed.
+The EgoSpeak GRU A+V row is taken from the published five-seed Ego4D
+table \cite{kim2025egospeak}; it is not a paired rerun under our
+causal-valid protocol. EasyCom SVA has one four-epoch seed-0 run, and Ego4D
+$\G+\A+$SVA has one measured run on the same 87 test clips; matched PTR
+remains unmeasured.
 
 \textbf{Metrics.}
-At each horizon we compute AP for the three activity channels over common
+Following EgoSpeak \cite{kim2025egospeak}, we compute average precision (AP)
-causal-valid rows, then average classes to obtain mAP and average all ten
+separately for background, wearer, and others at each horizon over common
-horizons to obtain mean mAP.  We also report wearer AP at 0.2~s.  For endpoint
+causal-valid cutoff--target pairs. Their unweighted class average is the
-diagnostics, onset AP conditions source rows on the wearer being silent and
+horizon mAP; Mean averages mAP over the ten horizons from 0.2 to 2.0~s.
-scores future speech with $p(\mathrm{wearer})$; offset AP conditions on wearer
+All reported AP and mAP values are percentages.
-speech and scores future silence with $1-p(\mathrm{wearer})$.  These
-threshold-free quantities average ten horizons.  They are not one-event-one-
-vote detection, semantic ENTRY/YIELD, or evidence of addressee understanding.
 
 \section{Results and Analysis}
 \label{sec:results}
 
-\subsection{EasyCom: Local Encoder Evidence}
+\noindent\textbf{Dense activity forecasting (Table~\ref{tab:main}).}
+All eight curves decline with horizon. On EasyCom, seed-0 $\G+\A$ scores
+81.72/65.11/56.16\% at 0.2~s/ten-horizon mean/2.0~s, versus
+73.45/60.85/54.32\% for $\G$ and 74.20/61.19/54.82\% for released
+EgoSpeak A+V. The $\G+\A$ gap over $\G$ narrows from 8.27 to 1.84
+points across the range, with 4.46 points remaining at 1.0~s. A separate audio-only screen scores
+79.69/60.51\% at 0.2~s/mean; differing capacity and history make fusion
+gains descriptive. On Ego4D, paired seeds 0--2 give $\G+\A$
+80.78/70.67/65.55\% versus $\G$ 74.72/67.56/63.98\%; the published
+EgoSpeak row gives 76.4/68.2/64.4\% under a different protocol. The paired
+$\G+\A$ margin is 2.85 points at 1.0~s and 1.57 at 2.0~s.
+The $\G+\A$ row remains above $\G$ at all ten horizons on both datasets.
+Because these tables report mAP rather than classwise AP, they do not
+identify whether background, wearer, or others drives each gain.
 
-Table~\ref{tab:main} anchors the measured EasyCom comparison. In a separate
+\noindent\textbf{Visual and privileged evidence
-audio-only screen, recent encoder-side $\A$ reaches 79.69\% at 0.2~s and
+(Tables~\ref{tab:main} and~\ref{tab:visual}).}
-60.51\% mean mAP; integrated $\G$ is slightly stronger on the ten-horizon
+EasyCom Pool/SVA/TSVA score 82.03/65.40, 82.36/65.37, and
-mean. Their seed-0 fusion gains 4.60 mean-mAP points over $\A$ and 4.26
+82.38/65.54\% at 0.2~s/mean, versus 81.72/65.11\% for $\G+\A$.
-over $\G$. Relative to released EgoSpeak, $\G+\A$ gains 7.52 points at
+Table~\ref{tab:main} shows only a 0.04-point SVA advantage at 2.0~s.
-0.2~s and 3.92 mean-mAP points.  Raw wearer offset AP rises by 7.23
+At 1.0~s, EasyCom SVA scores 63.73\% versus 63.81\% for $\G+\A$,
-points, onset AP by only 1.47: dense gains do not imply equal improvement
+so the visual difference is not uniform over horizons.
-at both activity boundaries.
+Ego4D's single SVA run scores 81.39/70.93\% against the three-seed
+$\G+\A$ mean of 80.78/70.67\%, but differs by only 0.01 points at
+2.0~s. Head/training differences prevent isolating a visual effect.
+In Table~\ref{tab:visual}, privileged visibility-only and one-second-delayed
+activity score 81.95/65.22 and 81.95/65.34\%, while timely visible OR
+scores 86.70/66.95\%. Its 4.75-point near-horizon and 1.73-point mean
+gaps over visibility-only suggest useful activity timing, but transcript
+labels are unavailable at deployment and are not an upper bound. None of
+these screens establishes speaker grounding or fewer false wearer entries.
+Fig.~\ref{fig:hq_cases} shows two outcome-selected 0.6-s Pool traces, not a
+population-level or modality-specific gain.
 
-The $\G$-only head has more trainable parameters and longer history than the
+\noindent\textbf{Context and chunk construction
-$\A$-only head. Thus this screen is consistent with predictive
+(Tables~\ref{tab:stream_context} and~\ref{tab:chunk}).}
-complementarity, not a capacity-matched proof that the Thinker lacks a
+The $\G+\A$ context-cap screen gives 64.81/65.18/64.74\% mean mAP at
-specific local cue. Individual $\A$ and $\G$ horizon curves are unavailable
+4/8/12~s, a 0.44-point range without a monotonic trend. The 8-s bank
-under the same causal-valid summary and are not reconstructed in Table~\ref{tab:main}.
+is 0.37/0.44 points above 4/12~s, but predates joint extraction and no
+matched $\G$ cap arm exists. In the
+separate global-only chunk screen, native hold scores 59.62/55.19\% at
+0.2~s/mean with 9,539 processor calls; non-native 200-ms commits score
+67.05/57.84\% with 95,453 calls. The observed-tail row scores
+70.85/59.17\%, with call count unavailable. Commits use roughly ten times
+as many processor calls as native hold; this is not a latency ratio.
+Changed chunk construction
+and unmatched extraction preclude an isolated PTR or throughput claim.
 
-% Retained as a planning aid; illustrative, unmeasured patterns do not belong
+\noindent\textbf{Backbone and claim limits (Table~\ref{tab:backbone}).}
-% in the page-limited submission preview.
+With a single-final-state readout, 3B $\G/\G+\A$ score 58.04/62.20\%
-\iffalse
+(+4.16 points); 7B score 57.70/62.18\% (+4.48). This seed-0 interface
-\begin{table*}[t]
+differs from the four-layer main readout and does not rank backbone sizes.
-\caption{Six-group \textbf{illustrative scenario, not measured results}.
+Across all tables, AP/mAP assess future activity rather than semantic
-These are conditional, pre-run interpretations for the pending matrix, not
+floor-control decisions.
-EasyCom scores.  $\Delta$ denotes paired mean-mAP change on common valid
-cutoffs; signs and shape must be replaced by measured values even if contrary
-to the story.  L/T/C/B require seeds 0--2 and per-session reporting.}
-\label{tab:scenario}
-\centering
-\normalsize
-\renewcommand{\arraystretch}{1.12}
-\setlength{\tabcolsep}{4.2pt}
-\begin{tabular}{@{}p{0.095\textwidth}p{0.29\textwidth}p{0.29\textwidth}p{0.235\textwidth}@{}}
-\toprule
-Group & Frozen comparison & Illustrative pattern \emph{if observed} & Falsifier / interpretation limit \\
-\midrule
-L1--4 & $\A,\G,\G+\A,\G+\A+\V$; matched-width $\G$;
-past-delayed local cues; current-state persistence
-& $\Delta(\G+\A-\G)>0$ after width matching; delayed $\A$ loses
-near-horizon benefit; persistence does not recover the gain
-& Width-only or persistence matches fusion, or delay does not hurt:
-do not attribute gain to timely distinct evidence. \\
-\addlinespace[2.5pt]
-T1--2 & 2-s completed-only vs 200-ms observed-tail refresh,
-crossed with $\A$ on/off; causal replay and phase audit
-& Tail helps most without $\A$; local $\A$ still adds at each refresh;
-future-only media never changes prior readouts
-& No independent local gain under tail refresh implies redundancy;
-replay/leakage failure invalidates timing claims. \\
-\addlinespace[2.5pt]
-C1--2 & Genuine 2/4/6/8/10/12-s feature extraction; fixed 1-s $\A$;
-add $\G$-only at each cap for a window-stability claim
-& Fused performance plateaus around mid-range caps while
-$\G$-only remains more cap-sensitive
-& If long history substitutes for $\A$, local-evidence claim narrows;
-without C2, no robustness comparison is justified. \\
-\addlinespace[2.5pt]
-B0--2 & Paired $\G$ and $\G+\A$ on Qwen 3B/7B and, after causal
-interface audit, MiniCPM-o 4.5
-& Positive within-backbone paired $\Delta$ on all admissible rows;
-absolute backbone ranking is not inferred
-& Failed causal/history match is unsupported, not zero gain;
-one negative paired row defeats universal transfer. \\
-\addlinespace[2.5pt]
-H1/Q1 & Predefined onset/offset and other-pause/overlap slices;
-same-horizon, cutoff-marked case selection
-& Offset gains coexist with hard wearer-onset negatives; examples
-explain, rather than replace, counted slice statistics
-& Pure-negative slice AP is undefined; selected examples cannot
-establish reduced false-entry rate. \\
-\addlinespace[2.5pt]
-S1 & End-to-end p50/p95 latency and peak memory, including I/O,
-encoding, tail recomputation, local path, and boundaries
-& A feasible throughput--quality tradeoff can be plotted
-against 200-ms output cadence
-& If p95 exceeds cadence, describe logical 5-Hz prediction only,
-not real-time deployment. \\
-\bottomrule
-\end{tabular}
-\end{table*}
-\fi
-
-\subsection{EasyCom: Visual Evidence and Ablations}
-
-Table~\ref{tab:visual} separates visual screens from the cross-dataset main
-matrix. Adding Pool to $\G+\A$ produces only +0.304 points at 0.2~s and
-+0.291 mean mAP.  Its wearer increment is +0.020 at 0.2~s and +0.061 when
-averaged over horizons; onset/offset changes are +0.059/+0.120.  A richer
-current-frame person-set model is stable across seeds but reaches lower mean
-mAP than the historical pooled row under a non-matched selection protocol.
-These results do not support a claim of learned speaker grounding. A separate
-seed-0 no-box full-frame TSVA screen reaches 82.38\% at 0.2~s and 65.54\%
-mean mAP, versus 81.72/65.11\% for the existing $\G+\A$ row under the same
-causal-valid dense protocol. TSVA adds a two-frame 2-FPS token history and a
-shared GRU, while head capacity and training batch differ; raw boundary
-metrics and repeat seeds are unavailable. Thus the +0.66/+0.43-point
-difference does not isolate visual motion or floor behavior.
-
-The lower block of Table~\ref{tab:visual} is an analysis-only audit that
-substitutes two seconds
-of transcript-derived, strictly past visible-person activity on the 5-Hz
-grid. The displayed conditions test visibility alone, a one-second delay,
-and OR-pooled activity. These privileged labels are
-neither deployable visual features nor a performance upper bound.
-The three-seed audio baseline is 81.87\% at 0.2~s and 65.18\% mean mAP;
-aggregate visible activity improves on it by 4.826 and 1.775 points,
-respectively.
-Visibility alone changes mean mAP by only +0.040, and
-one-second-delayed activity by +0.164.  Thus availability or extra parameters
-cannot explain the oracle gain.  Per-person activity exceeds shuffled identity
-by only 0.200 mean-mAP points and is itself 0.291 below the aggregate
-condition.  The current target therefore supports timely visible-floor
-activity as useful information, but not the necessity of identity continuity.
-The majority of the oracle's 0.2-s macro gain comes from background and
-others; wearer AP rises by only 0.572.  This is why neither the oracle nor a
-dense mAP gain establishes fewer false wearer entries.
-
-\begin{figure}[t]
-\centering
-\fbox{\begin{minipage}[c][18mm][c]{0.94\columnwidth}
-\centering
-\textbf{Q1 case slots: examples pending}\\[2pt]
-Onset / other-pause hard negative \quad|\quad Offset / overlap\\
-Show source cutoff, observed activity and same-horizon wearer scores.
-\end{minipage}}
-\caption{Reserved H/Q case figure. Select examples by a frozen rule after
-the slice counts in Table~\ref{tab:hard_slices} are available. No case or
-behavioral improvement is asserted by this placeholder.}
-\label{fig:hq_cases}
-\end{figure}
-
-\subsection{Chunk Timing and Cross-Dataset Limits}
-
-Table~\ref{tab:chunk} reports a separate global-only
-chunk-construction screen.  Non-native 200-ms commits exceed native 2-s hold
-at 0.2~s (67.05 versus 59.62\% mAP), but use 95,453 rather than 9,539
-processor calls over 12 sessions.  Completed blocks plus observed tail reach
-70.85\%; that older bank is not extraction-matched.  Freshness and chunk
-construction vary together, so this is motivation for, not evidence of, an
-isolated PTR benefit or real-time throughput. Table~\ref{tab:stream_context}
-reserves the matched tail $\times$ audio and context-cap tests.
-Tables~\ref{tab:backbone}--
-\ref{tab:efficiency} reserve backbone, hard-slice, and cost analyses; the
-cutoff-marked H/Q case in Fig.~\ref{fig:hq_cases} remains unmeasured.
-\textbf{Historical LoRA diagnostic.}
-A single-run EasyCom 8-s/2-FPS screen used sparse training anchors (1~Hz,
-plus 5~Hz near transitions) and dense 5-Hz evaluation.  Within that legacy
-pipeline, strict mean mAP was 58.17\% for a frozen dense-cache control,
-57.21\% for frozen sparse training, and 56.21\% for a separate three-epoch
-Thinker-LoRA run initialized from a trained frozen head.  An epoch-1 LoRA
-weights-only continuation fell to 54.21\% after optimizer/scheduler reset.
-Sparse frozen training is 0.96 points below the dense control; the separate
-LoRA run is another 1.01 below sparse frozen, but the epoch budgets differ.
-Sparse-train/dense-test mismatch and the optimizer reset are possible, not
-isolated, explanations.
-Missing remote raw metrics, matched seeds, and feature equivalence make this
-a negative recipe diagnostic, not a Table~\ref{tab:main} $\G+\A$ comparison
-or proof that LoRA is inherently harmful.
-Ego4D has no archived test metrics in this checkout, so transfer cannot yet
-be claimed.  The oracle uses privileged labels, provided boxes omit online
-detection, and aggregate others cannot establish addressee or semantic floor
-ownership.
-
-\subsection{Decision Rules and Reproducibility}
-
-Matched comparisons require identical cutoffs and targets; archive source
-timestamps, valid-pair counts, seeds, per-session AP, trainable counts, and
-checkpoint/processor revisions.  The fixed recipe and selection rules use
-training-side evidence only.  Already inspected test sessions make repeat
-seeds exploratory.  Raw endpoint AP is not one-event-one-vote ENTRY/YIELD:
-response windows must end at the next other onset, and yield windows at the
-next other offset.  Behavioral claims await a real-timeline evaluator audit.
 
 \section{Conclusion}
 
 For egocentric when-to-speak forecasting, cutoff-aligned access to recent
 encoder evidence is a useful complement to frozen Omni context on EasyCom;
-the strongest measured boundary difference is at wearer speech offset.
+the measured mAP separation is largest near the prediction cutoff.
 Pooled vision recovers little of the timely visible-activity information
 revealed by the privileged audit. PTR supplies a causal 5-Hz readout rule,
 but its independent predictive and efficiency effects still require matched
-tests. Ego4D transfer and censored event evaluation remain necessary before
+tests. The Ego4D screen repeats the dense audio gain under its own protocol, with a small
-claiming generalizable or improved semantic floor behavior.
+descriptive SVA gain in one run; censored event evaluation remains
-
+necessary before claiming generalizable or improved semantic floor
-\section{Acknowledgment}
+behavior. These results therefore support dense future-activity prediction,
-OpenAI Codex assisted with drafting and figure layout; the authors remain
+while the visual screens remain descriptive and the transcript-derived
-responsible for all experimental claims and final content.
+activity audit remains an information probe. A decisive test must hold
+cutoffs, context caps, training recipe, and valid pairs fixed while crossing
+PTR with local audio. It should then report per-session variability and
+event-level false entries and yields after censoring competing speaker
+changes. Only then can forecast gains support claims about an agent's floor
+decisions.
 
 \clearpage
 \bibliographystyle{IEEEbib}

manuscript/targeted/tables_compact.tex

@@ -1,11 +1,9 @@
 % Active result tables. A dash denotes an unreported or unmeasured value.
 \begin{table*}[t]
-\caption{Dense future-activity mAP (\%) at ten horizons on EasyCom and
+\caption{Ten-horizon future-activity mAP (\%). EasyCom Omni rows are
-Ego4D. Each dataset has the same four-method comparison. EasyCom values
+seed-0 screens; Ego4D $\G$/$\G+\A$ are three-seed means, and SVA is one
-are measured on Sessions~1--3; $\G$ has only archived 0.2-s and mean
+run. The published EgoSpeak Ego4D row \cite{kim2025egospeak} uses a
-scores, and SVA awaits evaluation. Ego4D scores are pending. Dashes are unreported or unmeasured,
+different protocol; cross-protocol differences are not paired effects.}
-not zero; intermediate scores are never interpolated. The available Omni
-rows are exploratory and not all training-matched.}
 \label{tab:main}
 \centering\ninept
 \renewcommand{\arraystretch}{0.91}
@@ -17,25 +15,24 @@ Dataset / model & .2 & .4 & .6 & .8 & 1.0 & 1.2 & 1.4 & 1.6 & 1.8 & 2.0 & Mean \
 \midrule
 \multicolumn{12}{@{}l}{\emph{EasyCom}} \\
 EgoSpeak A+V & 74.20 & 68.35 & 64.31 & 61.13 & 59.77 & 58.81 & 57.50 & 57.09 & 55.95 & 54.82 & 61.19 \\
-Omni $\G$ & 73.45 & -- & -- & -- & -- & -- & -- & -- & -- & -- & 60.85 \\
+Omni $\G$ & 73.45 & 68.64 & 64.65 & 61.51 & 59.35 & 58.08 & 57.22 & 56.15 & 55.12 & 54.32 & 60.85 \\
 Omni $\G+\A$ & 81.72 & 75.92 & 70.38 & 66.46 & 63.81 & 61.37 & 60.00 & 58.32 & 56.95 & 56.16 & 65.11 \\
-Omni $\G+\A+\mathrm{SVA}$ & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- \\
+Omni $\G+\A+\mathrm{SVA}$ & 82.36 & 76.26 & 70.89 & 66.67 & 63.73 & 61.52 & 60.10 & 58.63 & 57.35 & 56.20 & 65.37 \\
 \midrule
 \multicolumn{12}{@{}l}{\emph{Ego4D}} \\
-EgoSpeak A+V & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- \\
+EgoSpeak GRU A+V & 76.4 & 73.0 & 70.4 & 68.5 & 67.1 & 66.3 & 65.6 & 65.2 & 64.7 & 64.4 & 68.2 \\
-Omni $\G$ & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- \\
+Omni $\G$ & 74.72 & 72.03 & 69.81 & 68.11 & 66.81 & 65.88 & 65.21 & 64.73 & 64.35 & 63.98 & 67.56 \\
-Omni $\G+\A$ & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- \\
+Omni $\G+\A$ & 80.78 & 77.03 & 73.88 & 71.49 & 69.66 & 68.31 & 67.33 & 66.63 & 66.05 & 65.55 & 70.67 \\
-Omni $\G+\A+\mathrm{SVA}$ & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- & -- \\
+Omni $\G+\A+\mathrm{SVA}$ & 81.39 & 77.41 & 74.35 & 71.73 & 69.61 & 68.43 & 67.47 & 66.91 & 66.48 & 65.56 & 70.93 \\
 \bottomrule
 \end{tabular*}
 \end{table*}
 
 \begin{table}[t]
-\caption{EasyCom local-vision and information-audit screens (\%). The
+\caption{EasyCom vision screens and privileged activity audit (\%).
-upper rows are exploratory encoder-based runs; the lower rows use
+Upper rows are exploratory encoder-feature runs; lower rows use
-non-deployable past activity (three-seed mean $\pm$ SD). Their training
+transcript-derived past activity unavailable at deployment (three-seed
-protocols differ, so cross-section differences are not matched effects.
+means). Cross-block differences are not matched model effects.}
-Branch names are defined in Sec.~\ref{sec:method}.}
 \label{tab:visual}
 \centering\ninept
 \renewcommand{\arraystretch}{0.92}
@@ -46,45 +43,37 @@ Condition & mAP$_{.2}$ & Mean \\
 \midrule
 $\G+\A$ & 81.72 & 65.11 \\
 $\G+\A+$Pool & 82.03 & 65.40 \\
-$\G+\A+$SVA & -- & -- \\
+$\G+\A+$SVA & 82.36 & 65.37 \\
 $\G+\A+$TSVA & 82.38 & 65.54 \\
 \midrule
-Visibility only & $81.95{\pm}.10$ & $65.22{\pm}.05$ \\
+Visibility only & 81.95 & 65.22 \\
-Delayed 1 s & $81.95{\pm}.09$ & $65.34{\pm}.05$ \\
+Delayed 1 s & 81.95 & 65.34 \\
-Visible OR & $\mathbf{86.70{\pm}.09}$ & $\mathbf{66.95{\pm}.02}$ \\
+Visible OR & $\mathbf{86.70}$ & $\mathbf{66.95}$ \\
 \bottomrule
 \end{tabular}
 \end{table}
 
 \begin{table}[t]
-\caption{Pending matched timing/context tests. Each cell will compare
+\caption{EasyCom PTR context-cap screen: three-seed mean mAP (\%).
-$\G$ with $\G+\A$ on common cutoffs; $\Delta$ is paired mean-mAP gain.
+The 8-s bank predates joint extraction; no matched $\G$ arm exists.}
-PTR uses completed blocks plus the observed tail. Dashes are unmeasured.}
 \label{tab:stream_context}
 \centering\ninept
 \renewcommand{\arraystretch}{0.92}
 \setlength{\tabcolsep}{3pt}
-\begin{tabular}{@{}llcc@{}}
+\begin{tabular}{@{}cc@{}}
 \toprule
-Group & Refresh / cap & $\G/\G+\A$ & $\Delta$ \\
+PTR context cap (s) & $\G+\A$ mean mAP \\
 \midrule
-T & Completed hold, 8 s & --/-- & -- \\
+4 & 64.81 \\
-T & PTR, 8 s & --/-- & -- \\
+8 & 65.18 \\
-\midrule
+12 & 64.74 \\
-C & PTR, 2 s & --/-- & -- \\
-C & PTR, 4 s & --/-- & -- \\
-C & PTR, 6 s & --/-- & -- \\
-C & PTR, 8 s & --/-- & -- \\
-C & PTR, 10 s & --/-- & -- \\
-C & PTR, 12 s & --/-- & -- \\
 \bottomrule
 \end{tabular}
 \end{table}
 
 \begin{table}[t]
-\caption{Separate seed-0 global-only chunk screen (\%): 10 epochs, 8-s cap,
+\caption{Separate seed-0 global-only chunk screen (\%). PTR is not
-2-FPS video. PTR is not extraction-matched to the other rows. Calls are
+extraction-matched; calls are processor invocations, not latency.}
-processor invocations across 12 sessions, not latency.}
 \label{tab:chunk}
 \centering\ninept
 \renewcommand{\arraystretch}{0.95}